INDEX

Explanations

proper nouns related to military or historical figures

oai_token-act-pair · gpt-3.5-turbo

references to specific characters and themes related to monks

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

alez

-1.02

 horizont

-0.85

IUM

-0.80

ãĥ¤

-0.79

ocial

-0.73

ortmund

-0.72

acio

-0.71

ators

-0.70

itional

-0.68

itaire

-0.67

POSITIVE LOGITS

ucket

0.93

letcher

0.81

aughs

0.75

mare

0.73

mares

0.72

ucky

0.72

 Phant

0.68

OWN

0.68

orld

0.68

 Fletcher

0.68

Activations Density 0.064%

proper nouns related to military or historical figures

references to specific characters and themes related to monks

No Known Activations