INDEX

Explanations

terms related to conflict and controversy

oai_token-act-pair · gpt-3.5-turbo

terms related to conflict or division

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

spect

-0.77

racuse

-0.74

elect

-0.73

ulators

-0.70

ropolitan

-0.68

ascript

-0.68

electric

-0.67

gain

-0.67

opian

-0.66

ifax

-0.66

POSITIVE LOGITS

erers

0.80

bare

0.74

emonium

0.70

AGES

0.68

erer

0.68

 Flesh

0.66

 Bauer

0.66

AX

0.66

Reviewed

0.65

 Missions

0.64

Activations Density 0.101%

terms related to conflict and controversy

terms related to conflict or division

No Comments

No Known Activations