INDEX

Explanations

numbers related to voting or popularity rankings

oai_token-act-pair · gpt-3.5-turbo

numerical data or statistics related to votes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

holder

-0.80

istically

-0.71

leck

-0.71

anke

-0.71

uyomi

-0.67

think

-0.67

owicz

-0.67

pper

-0.66

ging

-0.66

 depos

-0.65

POSITIVE LOGITS

 ILCS

1.00

dB

0.95

0.86

mph

0.85

é¾

0.81

dB

0.80

0.78

0.75

0.74

Activations Density 0.015%

numbers related to voting or popularity rankings

numerical data or statistics related to votes

No Comments

No Known Activations