INDEX

Explanations

words related to visual perception, specifically the verb "saw."

oai_token-act-pair · gpt-3.5-turbo

instances of the verb "saw" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

 agric

-0.74

currency

-0.70

xus

-0.65

ÃŃa

-0.63

cult

-0.62

 isot

-0.62

adding

-0.61

 mantle

-0.61

 fridge

-0.61

vernment

-0.61

POSITIVE LOGITS

dust

1.13

lights

0.98

saw

0.81

PsyNetMessage

0.80

yers

0.80

awed

0.80

 firsthand

0.78

Ĺ

0.76

lasses

0.76

netflix

0.75

Activations Density 0.070%

words related to visual perception, specifically the verb "saw."

instances of the verb "saw" in various contexts

No Comments

No Known Activations