INDEX

Explanations

proper nouns related to locations

oai_token-act-pair · gpt-3.5-turbo

the presence of the term "Col" followed by a number, likely indicating geographical or organizational references

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Werewolf

-0.75

 Remastered

-0.74

natureconservancy

-0.70

 forth

-0.69

cffffcc

-0.64

 Ducks

-0.64

DragonMagazine

-0.62

 Icelandic

-0.62

visor

-0.62

 lockout

-0.61

POSITIVE LOGITS

ossal

1.26

onel

1.19

umbo

1.07

ours

1.04

ombo

1.03

onial

1.00

leen

1.00

oured

0.96

ossus

0.93

ouring

0.93

Activations Density 0.008%

proper nouns related to locations

the presence of the term "Col" followed by a number, likely indicating geographical or organizational references

No Comments

No Known Activations