INDEX

Explanations

words related to characteristics or features that are described or defined based on a particular concept or element

oai_token-act-pair · gpt-3.5-turbo

phrases that include the concept of "based" or "base" in various contexts

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.6.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

IVER

-0.82

 Niagara

-0.78

sen

-0.76

IVERS

-0.76

osen

-0.72

older

-0.70

 Wend

-0.70

OUGH

-0.70

LV

-0.70

nel

-0.68

POSITIVE LOGITS

 therapies

0.93

 solutions

0.86

 structures

0.83

 systems

0.80

 crowdfunding

0.80

 economy

0.80

 interventions

0.79

 strategies

0.79

 harassment

0.77

 aggress

0.77

Activations Density 0.029%

words related to characteristics or features that are described or defined based on a particular concept or element

phrases that include the concept of "based" or "base" in various contexts

No Comments

No Known Activations