INDEX

Explanations

proper names, specifically the name "Andrea"

oai_token-act-pair · gpt-3.5-turbo

occurrences of the name "Andrea" and related variations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

paced

-0.87

uilt

-0.82

lif

-0.81

quartered

-0.81

doms

-0.80

eer

-0.79

medium

-0.79

Ħ¢

-0.78

teen

-0.78

dress

-0.77

POSITIVE LOGITS

 Rossi

1.21

 Andrea

1.02

 Perez

0.91

 Grimes

0.86

 Dani

0.80

Ces

0.79

 Strauss

0.79

 Merkel

0.78

 Tyson

0.77

 Morse

0.76

Activations Density 0.011%

proper names, specifically the name "Andrea"

occurrences of the name "Andrea" and related variations

No Comments

No Known Activations