INDEX

Explanations

adjectives indicating importance or necessity

oai_token-act-pair · gpt-3.5-turbo

terms indicating necessity or importance

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

 blister

-0.69

respect

-0.66

 fever

-0.66

sheets

-0.64

ravings

-0.62

dust

-0.60

guyen

-0.59

robe

-0.57

rive

-0.57

 itch

-0.56

POSITIVE LOGITS

arna

0.80

arily

0.76

nell

0.76

 enough

0.69

uble

0.68

isSpecialOrderable

0.67

 nonetheless

0.67

ĸļ

0.65

ioned

0.65

edi

0.65

Activations Density 0.193%

adjectives indicating importance or necessity

terms indicating necessity or importance

No Comments

No Known Activations