INDEX

Explanations

phrases where a statement is emphasized or confirmed

oai_token-act-pair · gpt-3.5-turbo

repeated phrases indicating agreement or affirmation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

ĸļ

-0.69

arette

-0.65

ripp

-0.62

mat

-0.61

ains

-0.61

igmat

-0.60

ature

-0.60

ipation

-0.60

è¦ļéĨĴ

-0.60

odium

-0.59

POSITIVE LOGITS

eous

1.22

wing

0.79

 wing

0.79

move

0.76

shore

0.76

 winger

0.70

å¾

0.69

hand

0.68

mares

0.68

 fielder

0.68

Activations Density 0.053%

phrases where a statement is emphasized or confirmed

repeated phrases indicating agreement or affirmation

No Comments

No Known Activations