INDEX

Explanations

references to the public domain

oai_token-act-pair · gpt-3.5-turbo

terms related to the public domain and its usage

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

alos

-0.73

esta

-0.69

otos

-0.66

nar

-0.65

Blocks

-0.64

hillary

-0.63

athetic

-0.62

ophy

-0.61

Swe

-0.60

Scar

-0.59

POSITIVE LOGITS

 paperback

0.69

ensical

0.67

 dime

0.66

hower

0.63

 television

0.63

 dissemination

0.62

 depiction

0.60

 democracies

0.60

 journals

0.59

 pamph

0.59

Activations Density 0.150%

references to the public domain

terms related to the public domain and its usage

No Comments

No Known Activations