INDEX

Explanations

occurrences of the word "vote"

oai_token-act-pair · gpt-3.5-turbo

references to voting and legislative actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 millenn

-0.66

ensibly

-0.65

ahime

-0.65

iak

-0.64

 Andersen

-0.62

IJ

-0.62

Kut

-0.61

algia

-0.61

 pilgr

-0.61

hester

-0.61

POSITIVE LOGITS

 tally

0.90

 count

0.87

 vote

0.85

 unanimously

0.84

 votes

0.84

 approving

0.83

 booth

0.80

 authorizing

0.80

 bloc

0.78

rieve

0.78

Activations Density 0.029%

occurrences of the word "vote"

references to voting and legislative actions

No Comments

No Known Activations