INDEX

Explanations

phrases indicating directions or instructions

oai_token-act-pair · gpt-3.5-turbo

phrases related to taking action or making decisions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

leaders

-0.53

ãĥ¼ãĥĨãĤ£

-0.52

 prevailing

-0.52

ELD

-0.51

polit

-0.50

 Bureau

-0.50

åĲ

-0.50

Fair

-0.50

 Moody

-0.49

 frontline

-0.49

POSITIVE LOGITS

 masturb

0.92

 crochet

0.90

 uninstall

0.88

 orgasm

0.80

 download

0.77

 vape

0.76

 dont

0.76

 wanna

0.75

 browse

0.72

 realise

0.72

Activations Density 0.487%

phrases indicating directions or instructions

phrases related to taking action or making decisions

No Comments

No Known Activations