INDEX

Explanations

phrases related to physical attributes and actions

oai_token-act-pair · gpt-3.5-turbo

negative descriptions or characteristics of objects or situations

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

 Democracy

-0.72

national

-0.70

 Ethics

-0.70

iberal

-0.69

 Liberal

-0.68

democracy

-0.67

Government

-0.65

Liberal

-0.65

 Nation

-0.65

government

-0.64

POSITIVE LOGITS

 moisture

0.80

 texture

0.80

 waterproof

0.76

 textures

0.75

 noticeable

0.75

 rotate

0.74

 diagonal

0.74

 airflow

0.74

 tactile

0.73

 removable

0.73

Activations Density 2.265%

phrases related to physical attributes and actions

negative descriptions or characteristics of objects or situations

No Comments

No Known Activations