INDEX

Explanations

phrases related to titles with a specific pattern

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 @ 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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 unpre

-0.75

 lapse

-0.64

 doubling

-0.63

 Metallic

-0.63

 revived

-0.61

 flares

-0.61

 concerted

-0.59

 Prelude

-0.59

 bottleneck

-0.59

 reserve

-0.58

POSITIVE LOGITS

atson

1.39

alking

1.38

orst

1.37

restling

1.37

EEK

1.34

isdom

1.34

esley

1.32

ichita

1.29

atts

1.28

elsh

1.28

Activations Density 0.029%

phrases related to titles with a specific pattern

occurrences of the letter 'W' in various contexts

No Comments

No Known Activations