INDEX

Explanations

references to different models of products, such as electronic devices

oai_token-act-pair · gpt-3.5-turbo

references to specific product models and their variants

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

estern

-0.92

vernment

-0.88

pedia

-0.85

ulhu

-0.81

ipedia

-0.75

arching

-0.75

veyard

-0.73

apy

-0.72

isl

-0.72

arning

-0.72

POSITIVE LOGITS

 upgr

0.90

OEM

0.83

 manufactured

0.82

 upgrade

0.82

 variants

0.81

 manufacturer

0.78

 designation

0.78

kit

0.78

 sedan

0.76

 manufactures

0.76

Activations Density 0.137%

references to different models of products, such as electronic devices

references to specific product models and their variants

No Comments

No Known Activations