INDEX

Explanations

user comments or interaction related words

oai_token-act-pair · gpt-3.5-turbo

terms related to commenting and user interaction online

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Ã§a

-0.71

 Recon

-0.71

ccording

-0.68

nesses

-0.67

 Sinai

-0.65

pocket

-0.65

 turf

-0.65

lay

-0.64

Sensor

-0.64

 Lans

-0.63

POSITIVE LOGITS

ariat

1.02

ature

0.83

comments

0.83

aries

0.80

 comments

0.80

edin

0.79

ators

0.77

atively

0.77

 sections

0.74

 threads

0.74

Activations Density 0.040%

user comments or interaction related words

terms related to commenting and user interaction online

No Comments

No Known Activations