INDEX

Explanations

government-related terms and political actions

oai_token-act-pair · gpt-3.5-turbo

references to government and political processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

tics

-0.88

ãĤ¨ãĥ«

-0.81

oute

-0.73

iability

-0.73

ontent

-0.72

minus

-0.71

rosse

-0.71

Rank

-0.70

ounge

-0.69

 Baal

-0.68

POSITIVE LOGITS

 regulators

0.94

 legislation

0.94

 lawmakers

0.94

 mandated

0.92

 policymakers

0.88

 interven

0.85

 legisl

0.83

 intervened

0.83

 regulatory

0.83

 legislators

0.82

Activations Density 0.427%

government-related terms and political actions

references to government and political processes

No Comments

No Known Activations