INDEX

Explanations

mentions or descriptions of small or seemingly insignificant things

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "little."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

iership

-0.81

idents

-0.75

ï¸

-0.74

sem

-0.72

arching

-0.71

intent

-0.71

orial

-0.71

emale

-0.71

restling

-0.71

chwitz

-0.70

POSITIVE LOGITS

 girl

1.05

boy

0.99

 brother

0.94

 girls

0.91

 boys

0.91

guy

0.91

 sister

0.89

 helper

0.88

bit

0.87

 snippets

0.87

Activations Density 0.031%

mentions or descriptions of small or seemingly insignificant things

occurrences of the word "little."

No Comments

No Known Activations