INDEX

Explanations

names and references related to politics and specific individuals

oai_token-act-pair · gpt-3.5-turbo

proper nouns and names associated with individuals

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

itol

-0.72

©¶æ

-0.71

 crunch

-0.66

achev

-0.65

amic

-0.64

 artificially

-0.64

icipated

-0.62

ĳå£«

-0.60

 guiActiveUn

-0.60

estinal

-0.59

POSITIVE LOGITS

schild

1.02

orthy

0.78

conn

0.73

Redditor

0.73

orld

0.72

aii

0.72

merce

0.71

enegger

0.71

WARD

0.70

maker

0.68

Activations Density 0.234%

names and references related to politics and specific individuals

proper nouns and names associated with individuals

No Comments

No Known Activations