INDEX

Explanations

mentions of episodes in a series

oai_token-act-pair · gpt-3.5-turbo

mentions of "episodes" in relation to television shows or series

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

ntil

-0.85

achev

-0.74

cle

-0.72

 clipboard

-0.72

cule

-0.69

lly

-0.69

ILA

-0.67

pse

-0.67

 wheelchair

-0.66

ller

-0.66

POSITIVE LOGITS

 episodes

1.31

isodes

1.21

 episode

1.09

episode

0.98

ĸļ

0.89

 Episode

0.88

Episode

0.88

 DVDs

0.84

 "$:/

0.76

 aired

0.75

Activations Density 0.007%

mentions of episodes in a series

mentions of "episodes" in relation to television shows or series

No Comments

No Known Activations