INDEX

Explanations

references to news articles or reports

oai_token-act-pair · gpt-3.5-turbo

occurrences of square brackets

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

nesday

-0.75

 halfway

-0.73

 redu

-0.71

ores

-0.71

imb

-0.69

upl

-0.68

 comprom

-0.68

wagen

-0.67

imore

-0.66

 piping

-0.65

POSITIVE LOGITS

?]

1.40

!]

1.21

Laughs

1.20

Footnote

1.19

externalActionCode

1.19

emphasis

1.14

Pg

1.14

laughs

1.12

](

1.08

sic

1.06

Activations Density 0.027%

references to news articles or reports

occurrences of square brackets

No Comments

No Known Activations