INDEX

Explanations

proper nouns preceded by "Mr." (Mister)

oai_token-act-pair · gpt-3.5-turbo

titles or mentions of male characters

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

ãĥ¼ãĥĨ

-0.78

 actionGroup

-0.78

rawdownloadcloneembedreportprint

-0.69

anwhile

-0.69

 destro

-0.69

SHIP

-0.67

 damp

-0.64

 Compared

-0.62

 bars

-0.61

 wills

-0.61

POSITIVE LOGITS

 Universe

0.88

 Incredible

0.76

imen

0.72

angelo

0.71

 Claus

0.71

rahim

0.68

ude

0.67

gery

0.65

uv

0.65

ufact

0.65

Activations Density 0.035%

proper nouns preceded by "Mr." (Mister)

titles or mentions of male characters

No Comments

No Known Activations