INDEX

Explanations

words related to legal terms or criminal activities

oai_token-act-pair · gpt-3.5-turbo

words related to addiction and its various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

sie

-0.75

steen

-0.70

anas

-0.69

zee

-0.69

prints

-0.66

Kee

-0.65

thing

-0.62

 Schne

-0.62

chester

-0.61

DAQ

-0.61

POSITIVE LOGITS

aries

0.98

naire

0.98

iction

0.92

ictions

0.82

entious

0.77

urus

0.77

icult

0.75

ORD

0.75

ategory

0.74

inent

0.72

Activations Density 0.019%

words related to legal terms or criminal activities

words related to addiction and its various contexts

No Comments

No Known Activations