INDEX

Explanations

phrases that hint at or prompt for a response from the reader

oai_token-act-pair · gpt-3.5-turbo

expressions that prompt agreement or validation in conversation

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

itiz

-0.81

»Ĵ

-0.76

inki

-0.69

Ã¶r

-0.68

hurst

-0.67

aum

-0.66

ido

-0.65

opens

-0.64

icum

-0.62

icon

-0.62

POSITIVE LOGITS

?'

0.83

?!

0.82

?'"

0.81

?).

0.77

?".

0.76

?ãĢį

0.74

?]

0.74

????

0.74

!?"

0.73

Activations Density 0.086%

phrases that hint at or prompt for a response from the reader

expressions that prompt agreement or validation in conversation

No Comments

No Known Activations