INDEX

Explanations

terms, concepts, and definitions from various fields such as medicine, law, and technology

oai_token-act-pair · gpt-3.5-turbo

definitions or explanations of concepts

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

 mathemat

-0.81

 avenues

-0.78

 ports

-0.77

 bells

-0.76

 axes

-0.73

 lanes

-0.73

 distributions

-0.73

 channels

-0.73

 corridors

-0.72

 lines

-0.72

POSITIVE LOGITS

ogram

0.87

pload

0.82

raper

0.82

otomy

0.80

odge

0.77

ucket

0.77

ycle

0.77

ritic

0.75

ealous

0.75

igmatic

0.73

Activations Density 0.597%

terms, concepts, and definitions from various fields such as medicine, law, and technology

definitions or explanations of concepts

No Comments

No Known Activations