INDEX

Explanations

proper nouns related to locations, particularly those starting with "La" or "L"

oai_token-act-pair · gpt-3.5-turbo

references to specific people or entities, particularly those with the prefix "La."

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

MPH

-0.64

nyder

-0.63

poke

-0.63

 Junk

-0.59

 Bots

-0.57

PW

-0.57

 shenanigans

-0.55

 antis

-0.55

 talents

-0.54

 Rept

-0.54

POSITIVE LOGITS

rette

0.97

ĸļ

0.89

helle

0.77

anne

0.73

Ã±a

0.71

thia

0.70

ophon

0.70

otomy

0.69

anmar

0.69

uana

0.68

Activations Density 0.085%

proper nouns related to locations, particularly those starting with "La" or "L"

references to specific people or entities, particularly those with the prefix "La."

No Comments

No Known Activations