INDEX

Explanations

phrases indicating a clarification or further explanation

oai_token-act-pair · gpt-3.5-turbo

phrases that express personal opinions or invitations to listen

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

ibaba

-0.81

 accomp

-0.69

ombs

-0.69

 veter

-0.68

ibrary

-0.67

axter

-0.67

eatures

-0.66

 unden

-0.65

affiliated

-0.63

ioch

-0.63

POSITIVE LOGITS

lees

0.77

zzo

0.77

eeee

0.72

 thou

0.72

xff

0.69

 dear

0.67

yah

0.65

 Mister

0.65

sir

0.65

bro

0.64

Activations Density 0.094%

phrases indicating a clarification or further explanation

phrases that express personal opinions or invitations to listen

No Comments

No Known Activations