INDEX

Explanations

keywords related to television series and movie production, such as season numbers and production terms

oai_token-act-pair · gpt-3.5-turbo

references to seasons or episodic content in media

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

ertodd

-0.54

amina

-0.49

spin

-0.48

 preferably

-0.48

...]

-0.48

mble

-0.47

 paused

-0.47

orum

-0.46

Production

-0.45

 furthermore

-0.45

POSITIVE LOGITS

 predecessors

0.96

 predecessor

0.78

 previous

0.74

 counterparts

0.73

 earlier

0.59

 preceding

0.55

 eras

0.53

prev

0.52

 prior

0.52

 norm

0.52

Activations Density 1.187%

keywords related to television series and movie production, such as season numbers and production terms

references to seasons or episodic content in media

No Comments

No Known Activations