INDEX

Explanations

references to future plans or intentions

oai_token-act-pair · gpt-3.5-turbo

references to future intentions or strategies

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

wcsstore

-0.77

 Stain

-0.75

asions

-0.68

ruciating

-0.67

selves

-0.67

isha

-0.65

 Jude

-0.65

irrel

-0.64

 Chin

-0.63

mint

-0.61

POSITIVE LOGITS

 plans

1.14

 Plans

1.08

 rollout

0.85

paren

0.80

 reimburse

0.80

 lawy

0.76

 plan

0.75

 ambitions

0.73

plan

0.71

 screenings

0.71

Activations Density 0.025%

references to future plans or intentions

references to future intentions or strategies

No Comments

No Known Activations