INDEX

Explanations

years or specific dates

oai_token-act-pair · gpt-3.5-turbo

references to historical timestamps and significant events

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

olson

-0.77

pherd

-0.72

 zones

-0.67

eele

-0.66

ãģį

-0.66

hood

-0.64

Kos

-0.62

selves

-0.61

uyomi

-0.61

pb

-0.61

POSITIVE LOGITS

Via

0.96

ucci

0.86

 Racer

0.82

asca

0.74

hiba

0.72

coni

0.70

 Squadron

0.69

folio

0.68

igslist

0.68

 Fiat

0.68

Activations Density 0.002%

years or specific dates

references to historical timestamps and significant events

No Comments

No Known Activations