INDEX

Explanations

information related to movies and television shows

oai_token-act-pair · gpt-3.5-turbo

phrases related to movies and television shows

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

hirt

-0.57

eanor

-0.55

oway

-0.55

peed

-0.53

wit

-0.53

west

-0.53

enta

-0.52

 Maced

-0.51

veland

-0.50

arius

-0.50

POSITIVE LOGITS

rities

0.66

querque

0.63

dq

0.62

 Circus

0.61

etheus

0.54

apest

0.53

 Movies

0.53

ãĥŁ

0.53

 Engels

0.52

 movies

0.52

Activations Density 1.177%

information related to movies and television shows

phrases related to movies and television shows

No Comments

No Known Activations