INDEX

Explanations

the word "of"

oai_token-act-pair · gpt-3.5-turbo

the word "of."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

 prices

-0.72

 galleries

-0.69

 pled

-0.67

 corners

-0.66

 hordes

-0.65

 walls

-0.63

 partName

-0.62

 bidding

-0.62

 legions

-0.61

 riots

-0.61

POSITIVE LOGITS

rontal

1.16

onso

0.90

olia

0.89

icial

0.87

 course

0.87

omp

0.86

orm

0.85

ord

0.85

etheless

0.83

lower

0.83

Activations Density 0.032%

the word "of"

the word "of."

No Comments

No Known Activations