INDEX

Explanations

instances of words containing the sequence "elled"

oai_token-act-pair · gpt-3.5-turbo

words related to various forms of professional modeling or the term "modeled."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 disclosure

-0.73

Mun

-0.66

saf

-0.64

 hitch

-0.62

 security

-0.62

 Herb

-0.62

kn

-0.62

Ob

-0.60

 defense

-0.60

 vacation

-0.59

POSITIVE LOGITS

elled

4.55

eled

2.42

elling

2.14

ells

1.70

els

1.51

ELS

1.39

ell

1.35

eller

1.35

ellery

1.34

eling

1.32

Activations Density 0.007%

instances of words containing the sequence "elled"

words related to various forms of professional modeling or the term "modeled."

No Comments

No Known Activations