INDEX

Explanations

words related to political contexts and events

oai_token-act-pair · gpt-3.5-turbo

references to political entities and electoral processes

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

tics

-0.86

ãĤ¨ãĥ«

-0.75

ounge

-0.75

iability

-0.73

rosse

-0.72

station

-0.65

 Ascension

-0.65

utenberg

-0.65

ttes

-0.64

 Leilan

-0.64

POSITIVE LOGITS

 lawmakers

0.95

 approving

0.91

 legislators

0.90

 policymakers

0.89

 appoint

0.86

 approves

0.84

 legislation

0.83

 legisl

0.82

 initiatives

0.80

 legalize

0.79

Activations Density 0.392%

words related to political contexts and events

references to political entities and electoral processes

No Comments

No Known Activations