INDEX

Explanations

words related to maneuvers or maneuvers themselves

oai_token-act-pair · gpt-3.5-turbo

terms related to maneuvering and theatrical contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

 grip

-0.72

FUL

-0.72

nikov

-0.69

 Brach

-0.69

finger

-0.68

 Keeper

-0.67

ãĥīãĥ©ãĤ´ãĥ³

-0.66

ITE

-0.64

velength

-0.64

 Soup

-0.64

POSITIVE LOGITS

rative

1.27

rator

1.18

ration

1.04

rous

1.02

res

1.01

ered

1.01

rals

1.01

rates

1.00

rum

0.99

rament

0.98

Activations Density 0.022%

words related to maneuvers or maneuvers themselves

terms related to maneuvering and theatrical contexts

No Comments

No Known Activations