INDEX

Explanations

proper nouns related to specific locations or entities

oai_token-act-pair · gpt-3.5-turbo

references to specific names or titles, particularly those that include the word "Dos" or similar variations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

iste

-0.79

ists

-0.79

 reserve

-0.75

ograp

-0.72

isms

-0.71

icity

-0.70

ry

-0.69

OGR

-0.69

owner

-0.67

BG

-0.67

POSITIVE LOGITS

BuyableInstoreAndOnline

0.82

 Mons

0.78

peria

0.77

 Santos

0.74

illo

0.74

AGE

0.73

 Alto

0.72

eq

0.72

AGES

0.70

onest

0.69

Activations Density 0.115%

proper nouns related to specific locations or entities

references to specific names or titles, particularly those that include the word "Dos" or similar variations

No Comments

No Known Activations