INDEX

Explanations

proper nouns related to titles or names

oai_token-act-pair · gpt-3.5-turbo

the word 'compliant' and its variants

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

exc

-0.71

 Glas

-0.69

ear

-0.67

 Chal

-0.65

 repeated

-0.61

 every

-0.61

 repeat

-0.58

 Marx

-0.57

ass

-0.57

 revision

-0.57

POSITIVE LOGITS

iant

4.75

iants

2.88

iance

2.32

iances

1.60

iating

1.44

iate

1.40

iated

1.39

iation

1.32

iator

1.32

ient

1.30

Activations Density 0.008%

proper nouns related to titles or names

the word 'compliant' and its variants

No Comments

No Known Activations