INDEX

Explanations

phrases related to assigning importance or worth to something

oai_token-act-pair · gpt-3.5-turbo

expressions of value or valuation

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

nings

-0.77

hew

-0.74

 bankrupt

-0.63

bleacher

-0.62

thumbnails

-0.62

agnetic

-0.62

soType

-0.61

vic

-0.61

/**

-0.61

 wipes

-0.60

POSITIVE LOGITS

enance

0.93

 patience

0.85

lessly

0.84

 flexibility

0.83

ably

0.79

 liberty

0.77

 secrecy

0.76

 loyalty

0.76

 discretion

0.74

 Accuracy

0.73

Activations Density 0.121%

phrases related to assigning importance or worth to something

expressions of value or valuation

No Comments

No Known Activations