INDEX

Explanations

terms related to defense and legal proceedings

oai_token-act-pair · gpt-3.5-turbo

terms related to defense and military context

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

urat

-0.87

é¾įåĸļå£«

-0.81

olini

-0.71

acci

-0.70

iasco

-0.69

 grapes

-0.67

ragon

-0.67

 Ruler

-0.65

anny

-0.65

umn

-0.65

POSITIVE LOGITS

 Clapper

0.66

 Grimes

0.63

psi

0.62

Esk

0.62

 Dempsey

0.58

 Perception

0.58

ilities

0.57

atives

0.57

PsyNetMessage

0.56

ailability

0.56

Activations Density 0.426%

terms related to defense and legal proceedings

terms related to defense and military context

No Comments

No Known Activations