INDEX

Explanations

scientific or technical terms and proper nouns related to organizations

oai_token-act-pair · gpt-3.5-turbo

technical and structural terminology in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 Kell

-0.71

 Released

-0.71

 Kits

-0.70

dos

-0.69

 McGee

-0.68

Go

-0.68

Dos

-0.68

HS

-0.67

mc

-0.66

Cu

-0.66

POSITIVE LOGITS

livious

0.94

icism

0.93

ophon

0.91

antics

0.91

onic

0.90

otic

0.89

ativity

0.86

acular

0.86

functional

0.86

ulence

0.86

Activations Density 0.310%

scientific or technical terms and proper nouns related to organizations

technical and structural terminology in the text

No Comments

No Known Activations