INDEX

Explanations

terms related to deliberate, planned actions or events

oai_token-act-pair · gpt-3.5-turbo

words related to premeditation and intent

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

bye

-0.69

 turkey

-0.67

 Worldwide

-0.63

Ty

-0.63

 Zero

-0.60

OM

-0.59

Win

-0.58

 granted

-0.58

 Extension

-0.58

 mill

-0.56

POSITIVE LOGITS

itated

4.18

itate

3.91

itates

3.75

itating

3.63

itation

2.64

itations

2.46

itant

1.81

itative

1.79

itational

1.64

itous

1.56

Activations Density 0.018%

terms related to deliberate, planned actions or events

words related to premeditation and intent

No Comments

No Known Activations