INDEX

Explanations

words related to winning, losing, or competitiveness

oai_token-act-pair · gpt-3.5-turbo

terms associated with winning and losing, particularly in a competitive context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

ascript

-0.85

seeing

-0.84

ternity

-0.82

fi

-0.79

nery

-0.75

ledge

-0.75

osate

-0.74

iosis

-0.72

weed

-0.72

behind

-0.72

POSITIVE LOGITS

 streak

0.88

 professions

0.78

 bidder

0.74

 qualities

0.71

 edge

0.70

 mechanism

0.70

 heights

0.70

 Eleven

0.70

 factor

0.69

 percentages

0.69

Activations Density 0.206%

words related to winning, losing, or competitiveness

terms associated with winning and losing, particularly in a competitive context

No Comments

No Known Activations