INDEX

Explanations

names and terms related to various individuals and organizations

oai_token-act-pair · gpt-3.5-turbo

the recurrence of the word "and."

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 heck

-0.61

ichick

-0.56

 prominently

-0.56

 Tsukuyomi

-0.54

 Hawking

-0.53

PG

-0.53

 cann

-0.52

lda

-0.51

Story

-0.51

 ridicule

-0.51

POSITIVE LOGITS

ividual

1.17

ez

1.11

ication

1.02

elle

1.00

icate

0.99

icated

0.96

icates

0.94

ragon

0.93

orf

0.92

olin

0.91

Activations Density 0.023%

names and terms related to various individuals and organizations

the recurrence of the word "and."

No Comments

No Known Activations