INDEX

Explanations

words related to punk music and culture

oai_token-act-pair · gpt-3.5-turbo

references to punk culture and music

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

abet

-0.61

 exact

-0.60

rium

-0.59

 latitude

-0.58

Uncommon

-0.58

 Hubble

-0.58

uminum

-0.58

BUS

-0.57

 Lama

-0.57

++++++++++++++++

-0.57

POSITIVE LOGITS

punk

1.28

 punk

0.96

istani

0.87

ishly

0.81

 Punk

0.79

igans

0.76

ledge

0.75

acing

0.73

 rock

0.72

ettes

0.70

Activations Density 0.009%

words related to punk music and culture

references to punk culture and music

No Comments

No Known Activations