INDEX

Explanations

data and numerical values

oai_token-act-pair · gpt-3.5-turbo

numerical sequences and data representations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

 rhet

-0.75

 behav

-0.74

hyde

-0.67

ï¸

-0.64

 conduc

-0.62

 antit

-0.62

gently

-0.61

sonian

-0.61

 ingred

-0.61

bryce

-0.60

POSITIVE LOGITS

avg

0.76

etta

0.76

Avg

0.72

PHI

0.69

pees

0.67

 Saints

0.64

 Divinity

0.63

 Jagu

0.62

 Destiny

0.60

 Snapdragon

0.60

Activations Density 0.159%

data and numerical values

numerical sequences and data representations

No Comments

No Known Activations