INDEX

Explanations

concepts or actions related to mixing or blending different elements together

oai_token-act-pair · gpt-3.5-turbo

instances of the word "blend" and its variations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

dem

-0.70

selves

-0.66

Universal

-0.66

 negligent

-0.64

porary

-0.63

nai

-0.63

ockets

-0.62

pel

-0.62

 attendant

-0.61

ikhail

-0.60

POSITIVE LOGITS

 blends

1.30

 blend

1.25

 blending

1.02

 Flavoring

0.94

 Blend

0.93

 mixes

0.79

wana

0.78

 blended

0.73

 seamlessly

0.72

 Angelo

0.71

Activations Density 0.009%

concepts or actions related to mixing or blending different elements together

instances of the word "blend" and its variations

No Comments

No Known Activations