INDEX

Explanations

words related to valuable or historical items, particularly artifacts

oai_token-act-pair · gpt-3.5-turbo

terms related to historical artifacts and relics

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

rol

-0.87

nut

-0.78

yip

-0.78

rons

-0.74

kers

-0.74

ter

-0.73

tan

-0.72

creen

-0.72

sen

-0.68

yan

-0.68

POSITIVE LOGITS

ifacts

1.13

 arte

1.10

 artifacts

1.02

 artifact

1.00

 relics

0.93

artifacts

0.86

ifact

0.86

 Frie

0.77

 treasures

0.73

 relic

0.71

Activations Density 0.031%

words related to valuable or historical items, particularly artifacts

terms related to historical artifacts and relics

No Comments

No Known Activations