INDEX

Explanations

phrases indicating opposition or disagreement

oai_token-act-pair · gpt-3.5-turbo

phrases indicating conflict or disagreement between two entities

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ogle

-0.73

umper

-0.64

mson

-0.63

ãĤ®

-0.61

\<

-0.58

{*

-0.56

pletion

-0.55

adobe

-0.55

cens

-0.55

stad

-0.54

POSITIVE LOGITS

 respectively

1.83

 together

1.79

together

1.64

 mutually

1.63

 jointly

1.60

 alike

1.60

 Together

1.51

 interchange

1.47

 toget

1.35

 collide

1.32

Activations Density 1.116%

phrases indicating opposition or disagreement

phrases indicating conflict or disagreement between two entities

No Comments

No Known Activations