INDEX

Explanations

words related to experimental studies or research

oai_token-act-pair · gpt-3.5-turbo

references to experimental processes or studies

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

atra

-0.80

andra

-0.78

cript

-0.78

utra

-0.77

criptions

-0.77

holders

-0.76

veland

-0.74

kins

-0.73

pered

-0.73

adr

-0.73

POSITIVE LOGITS

imental

0.93

ists

0.89

izations

0.76

 Prototype

0.74

ization

0.72

oad

0.72

ising

0.71

ised

0.71

 collaborations

0.71

 explor

0.70

Activations Density 0.025%

words related to experimental studies or research

references to experimental processes or studies

No Comments

No Known Activations