INDEX

Explanations

terms related to heat or heating

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 Goff

-0.78

ournal

-0.74

ufact

-0.73

 Mandela

-0.71

ystem

-0.71

VICE

-0.69

vernment

-0.69

lus

-0.68

uthor

-0.66

rahim

-0.65

POSITIVE LOGITS

hens

1.36

wave

1.23

waves

1.17

seekers

1.15

stroke

1.08

 exch

1.06

hen

1.03

 exhaustion

0.97

seeking

0.87

 waves

0.86

Activations Density 0.043%

terms related to heat or heating

No Comments

No Known Activations