INDEX

Explanations

mentions of names or terms containing "Te" with other letters following them

oai_token-act-pair · gpt-3.5-turbo

specific characters or letters that are significant within a context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

 Andersen

-0.68

[+

-0.67

 liabilities

-0.62

 hostages

-0.59

 unsur

-0.58

 upkeep

-0.58

 wiser

-0.58

 indemn

-0.57

 elaborate

-0.56

 wrath

-0.56

POSITIVE LOGITS

ihu

0.95

reek

0.83

achus

0.76

akeru

0.75

illin

0.74

Tal

0.73

pter

0.72

inki

0.72

mouth

0.72

orage

0.72

Activations Density 0.109%

mentions of names or terms containing "Te" with other letters following them

specific characters or letters that are significant within a context

No Comments

No Known Activations