INDEX

Explanations

specific words or phrases placed "on" something

oai_token-act-pair · gpt-3.5-turbo

phrases involving the use of the word "on" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

MpServer

-0.65

imore

-0.63

 DRAG

-0.60

ugh

-0.60

 anticipated

-0.59

self

-0.59

ACE

-0.58

 VIDEOS

-0.57

ieth

-0.57

 Cosponsors

-0.53

POSITIVE LOGITS

tow

1.04

 them

0.84

tones

0.75

between

0.74

sole

0.73

itia

0.71

license

0.69

aband

0.68

deck

0.67

it

0.67

Activations Density 0.190%

specific words or phrases placed "on" something

phrases involving the use of the word "on" in various contexts

No Comments

No Known Activations