INDEX

Explanations

phrases related to legal or procedural statements

oai_token-act-pair · gpt-3.5-turbo

occurrences of the article "a"

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

lines

-0.69

lasses

-0.66

pter

-0.66

chart

-0.65

lins

-0.64

bite

-0.63

LIN

-0.62

erver

-0.62

 Orlando

-0.62

reports

-0.61

POSITIVE LOGITS

usterity

1.47

cknow

0.96

irst

0.95

esthetic

0.95

verages

0.88

 couple

0.85

ption

0.85

lot

0.84

qua

0.84

ird

0.83

Activations Density 0.082%

phrases related to legal or procedural statements

occurrences of the article "a"

No Comments

No Known Activations