INDEX

Explanations

mentions of rankings or positions of authority

oai_token-act-pair · gpt-3.5-turbo

instances of "ranking" or references to rankings in various contexts

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

ported

-0.72

berra

-0.71

RPG

-0.69

perty

-0.69

idem

-0.69

jab

-0.68

 Ezek

-0.67

fitting

-0.63

artney

-0.63

 fasting

-0.62

POSITIVE LOGITS

hyde

0.82

 ranking

0.81

Rank

0.78

enf

0.76

 citiz

0.73

 rankings

0.72

enance

0.71

essee

0.70

emouth

0.69

masters

0.69

Activations Density 0.011%

mentions of rankings or positions of authority

instances of "ranking" or references to rankings in various contexts

No Comments

No Known Activations