INDEX

Explanations

specific words related to selecting or choosing individual options

oai_token-act-pair · gpt-3.5-turbo

references to different types of communication channels

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

berman

-0.76

iqueness

-0.69

 Raiders

-0.68

hiro

-0.66

gets

-0.66

 Kear

-0.61

 Linda

-0.61

enhagen

-0.60

 McCl

-0.60

Gat

-0.59

POSITIVE LOGITS

 channels

0.95

 channel

0.92

lain

0.86

Channel

0.76

channel

0.75

ting

0.75

akra

0.71

nel

0.70

agall

0.69

izoph

0.69

Activations Density 0.013%

specific words related to selecting or choosing individual options

references to different types of communication channels

No Comments

No Known Activations