INDEX

Explanations

instances where something is being used for a specific purpose

oai_token-act-pair · gpt-3.5-turbo

phrases indicating purposes or uses of items or information

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

Collins

-0.72

ellen

-0.71

ttle

-0.69

Ohio

-0.67

*/

-0.66

fleet

-0.66

iev

-0.65

beat

-0.65

Bell

-0.64

idav

-0.64

POSITIVE LOGITS

 purposes

1.07

aging

0.93

agers

0.92

 storing

0.87

 constructing

0.87

 emergencies

0.86

 protection

0.86

 example

0.86

 comparison

0.85

bidden

0.84

Activations Density 0.085%

instances where something is being used for a specific purpose

phrases indicating purposes or uses of items or information

No Comments

No Known Activations