INDEX

Explanations

names of cities and dates

oai_token-act-pair · gpt-3.5-turbo

occurrences of the letter 'W' in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

 unpre

-0.82

 bottleneck

-0.74

 Prelude

-0.68

 gratification

-0.67

uate

-0.65

 Malfoy

-0.62

 apprehension

-0.62

 sucker

-0.62

Khe

-0.61

İĭ

-0.61

POSITIVE LOGITS

OW

1.25

restling

1.24

ITNESS

1.24

atts

1.22

ITCH

1.22

reck

1.21

alking

1.21

ALK

1.21

orthy

1.20

ORD

1.19

Activations Density 0.036%

names of cities and dates

occurrences of the letter 'W' in various contexts

No Comments

No Known Activations