INDEX

Explanations

technical terms related to data processing and procedures

oai_token-act-pair · gpt-3.5-turbo

technical terms and concepts related to systems, processes, and governance

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

 outright

-0.64

 Marketplace

-0.63

 Whitman

-0.62

 Breitbart

-0.61

 âĢķ

-0.61

 Petraeus

-0.61

 BuzzFeed

-0.60

 Aren

-0.59

Interested

-0.59

 Quinn

-0.58

POSITIVE LOGITS

 clauses

0.94

boards

0.92

posts

0.90

aires

0.88

ings

0.87

able

0.87

of

0.86

 plates

0.86

 sheets

0.85

cards

0.85

Activations Density 0.636%

technical terms related to data processing and procedures

technical terms and concepts related to systems, processes, and governance

No Comments

No Known Activations