INDEX

Explanations

phrases indicating lack of surprise

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

abouts

-0.84

abases

-0.67

tein

-0.64

inet

-0.63

heastern

-0.62

existent

-0.62

nels

-0.61

Administ

-0.60

peer

-0.59

nsic

-0.58

POSITIVE LOGITS

 surprise

1.25

 shock

0.96

 surprises

0.94

 Surprise

0.91

 surpr

0.90

 unexpected

0.87

 surprising

0.86

 disappointment

0.81

shock

0.81

 revelation

0.79

Activations Density 0.084%

phrases indicating lack of surprise

No Comments

No Known Activations