INDEX

Explanations

mentions of the word "Games"

oai_token-act-pair · gpt-3.5-turbo

references to video games

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 politic

-0.85

aye

-0.78

 acknow

-0.68

sie

-0.67

iffe

-0.67

 wildfire

-0.65

isin

-0.61

tow

-0.61

 annex

-0.59

 evac

-0.58

POSITIVE LOGITS

manship

1.31

 consoles

1.01

erver

1.01

cube

0.97

paces

0.93

wright

0.90

pad

0.88

pace

0.87

 Played

0.87

indust

0.85

Activations Density 0.030%

mentions of the word "Games"

references to video games

No Comments

No Known Activations