INDEX

Explanations

occurrences of the word "up" in various contexts

oai_token-act-pair · gpt-3.5-turbo

the phrase "show up" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

 Abyss

-0.65

fore

-0.64

 Whitman

-0.61

cheon

-0.60

 Bulg

-0.59

ï¿½ï¿½

-0.59

________________________

-0.59

Geh

-0.58

cknowled

-0.58

 mined

-0.58

POSITIVE LOGITS

dates

0.85

laws

0.80

law

0.80

votes

0.73

lier

0.73

dating

0.73

lander

0.69

CHAT

0.69

skirts

0.68

cycles

0.67

Activations Density 0.074%

occurrences of the word "up" in various contexts

the phrase "show up" in various contexts

No Comments

No Known Activations