INDEX

Explanations

proper names of individuals likely related to various fields or locations

oai_token-act-pair · gpt-3.5-turbo

references to individuals and entities associated with various activities or fields

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

İĭ

-0.84

 guarant

-0.72

 subp

-0.67

tsky

-0.65

 Cance

-0.61

apache

-0.61

 acron

-0.59

 disapp

-0.59

uyomi

-0.58

 carrot

-0.58

POSITIVE LOGITS

ette

0.74

inger

0.74

SHA

0.73

oola

0.72

cliffe

0.71

NB

0.70

ogle

0.70

Collins

0.70

ettes

0.67

Ã£

0.67

Activations Density 0.097%

proper names of individuals likely related to various fields or locations

references to individuals and entities associated with various activities or fields

No Comments

No Known Activations