INDEX

Explanations

links or connections between different concepts or variables

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate correlations or links between different subjects

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

quished

-0.96

OGR

-0.88

\\\\\\\\

-0.87

abases

-0.83

TPPStreamerBot

-0.78

stal

-0.76

////////////////

-0.76

spect

-0.75

iken

-0.75

scrib

-0.75

POSITIVE LOGITS

 these

0.71

 ethnicity

0.71

 criminality

0.71

 sexes

0.65

 counties

0.65

 dots

0.65

 humans

0.63

 academics

0.63

 disparate

0.63

 academia

0.62

Activations Density 0.040%

links or connections between different concepts or variables

phrases that indicate correlations or links between different subjects

No Comments

No Known Activations