INDEX

Explanations

the prefix "ins" or "Ins" in words

oai_token-act-pair · gpt-3.5-turbo

references to "insider" perspectives or insights

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

 Archdemon

-0.64

RTX

-0.61

 Baird

-0.60

Sao

-0.60

 Reconstruction

-0.60

 Sigma

-0.60

erous

-0.58

 Libraries

-0.57

 Remastered

-0.57

 McAuliffe

-0.57

POSITIVE LOGITS

urance

1.56

ufficient

1.56

omnia

1.53

pection

1.47

iders

1.46

urrection

1.43

ights

1.42

ensitive

1.42

ulin

1.42

ider

1.36

Activations Density 0.030%

the prefix "ins" or "Ins" in words

references to "insider" perspectives or insights

No Comments

No Known Activations