INDEX

Explanations

names of political figures or locations

oai_token-act-pair · gpt-3.5-turbo

words or abbreviations that represent names or organizations

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 PUBLIC

-0.72

enegger

-0.68

xp

-0.67

 fools

-0.67

BILITIES

-0.65

Fn

-0.64

plays

-0.63

+/-

-0.61

 SPACE

-0.60

 SHARES

-0.60

POSITIVE LOGITS

Hol

0.83

VP

0.79

rim

0.78

HR

0.75

igil

0.67

Nor

0.67

az

0.66

ams

0.66

west

0.66

 Stores

0.65

Activations Density 0.098%

names of political figures or locations

words or abbreviations that represent names or organizations

No Comments

No Known Activations