INDEX

Explanations

words related to actions and environments in a film setting

oai_token-act-pair · gpt-3.5-turbo

phrases or terms related to visual elements in media

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

 Everest

-0.78

 Majority

-0.73

 unlimited

-0.68

 incumbent

-0.67

 indefinite

-0.67

 Reserve

-0.66

 insurer

-0.66

 offshore

-0.66

 insured

-0.64

 unloaded

-0.64

POSITIVE LOGITS

isms

1.38

tones

1.37

notations

1.33

estyles

1.21

 animations

1.20

 phrases

1.19

 metaphors

1.16

 accents

1.15

 tones

1.15

actions

1.14

Activations Density 0.496%

words related to actions and environments in a film setting

phrases or terms related to visual elements in media

No Comments

No Known Activations