INDEX

Explanations

references to winning or victories

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "winning."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

heter

-0.71

agg

-0.67

thodox

-0.65

cht

-0.64

men

-0.63

trak

-0.63

den

-0.63

mu

-0.62

erity

-0.61

 footprint

-0.60

POSITIVE LOGITS

STON

0.73

 bidder

0.73

 victories

0.69

eteenth

0.67

 streak

0.66

escape

0.66

icent

0.66

now

0.65

't

0.65

boa

0.65

Activations Density 0.022%

references to winning or victories

occurrences of the word "winning."

No Comments

No Known Activations