INDEX

Explanations

mentions of the word "patents" in a document

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "patents."

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

ãĤ¨ãĥ«

-0.76

anchester

-0.74

lished

-0.70

ulative

-0.68

ebus

-0.63

dule

-0.62

elines

-0.62

udeb

-0.61

 Babel

-0.61

electric

-0.59

POSITIVE LOGITS

ents

1.08

ented

0.85

poons

0.85

lip

0.76

ent

0.73

enting

0.73

poon

0.73

ertodd

0.69

 indent

0.69

uits

0.67

Activations Density 0.007%

mentions of the word "patents" in a document

occurrences of the word "patents."

No Comments

No Known Activations