INDEX

Explanations

phrases related to investigations or crises

oai_token-act-pair · gpt-3.5-turbo

references to investigations, crises, and scandals involving political or legal matters

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 occasional

-0.61

poon

-0.60

 Dealer

-0.60

 consolation

-0.57

POV

-0.57

#$#$

-0.57

Live

-0.56

 pawn

-0.56

 duplicate

-0.56

 Converted

-0.56

POSITIVE LOGITS

 begun

0.83

 itself

0.82

iest

0.81

 unfolding

0.78

 underway

0.76

 referred

0.76

 outlined

0.74

 envis

0.74

 surrounding

0.73

Activations Density 0.371%

phrases related to investigations or crises

references to investigations, crises, and scandals involving political or legal matters

No Comments

No Known Activations