INDEX

Explanations

questions introduced with the word "What"

oai_token-act-pair · gpt-3.5-turbo

questions formulated with "What"

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

shore

-0.66

lich

-0.65

general

-0.64

roads

-0.64

ulic

-0.63

Lago

-0.63

ability

-0.63

println

-0.62

gi

-0.60

raction

-0.59

POSITIVE LOGITS

soever

1.24

 happens

1.04

 Lies

0.98

 happened

0.94

 distinguishes

0.94

 transpired

0.92

 Makes

0.90

 separates

0.88

 Difference

0.85

 happ

0.83

Activations Density 0.081%

questions introduced with the word "What"

questions formulated with "What"

No Comments

No Known Activations