INDEX

Explanations

words related to generalization or broad statements

oai_token-act-pair · gpt-3.5-turbo

words indicating generalization or vagueness in descriptions

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

quit

-0.79

rance

-0.78

chwitz

-0.74

 Ranger

-0.72

ername

-0.69

sic

-0.69

 chang

-0.69

gone

-0.66

DragonMagazine

-0.65

hitting

-0.64

POSITIVE LOGITS

 broadly

2.74

 loosely

2.40

 tightly

2.31

 narrowly

2.03

 evenly

1.83

 thinly

1.69

 neatly

1.48

 closely

1.44

 vaguely

1.44

 densely

1.42

Activations Density 0.020%

words related to generalization or broad statements

words indicating generalization or vagueness in descriptions

No Comments

No Known Activations