INDEX

Explanations

action words related to focus or dedication

oai_token-act-pair · gpt-3.5-turbo

prepositions and phrases indicating direction or focus

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

olitics

-0.88

ears

-0.79

udence

-0.76

omsky

-0.75

Mrs

-0.75

ISON

-0.73

icist

-0.72

ynes

-0.72

adh

-0.71

iannopoulos

-0.71

POSITIVE LOGITS

 integrating

1.69

 optimizing

1.67

 improving

1.63

 creating

1.60

 designing

1.55

 providing

1.53

 introducing

1.48

 incorporating

1.48

 delivering

1.47

 adapting

1.46

Activations Density 0.321%

action words related to focus or dedication

prepositions and phrases indicating direction or focus

No Comments

No Known Activations