INDEX

Explanations

dates, specifically in the format "Jun (day)"

oai_token-act-pair · gpt-3.5-turbo

dates and references to specific time periods

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

ngth

-0.96

Interstitial

-0.76

Ö¼

-0.71

 Trials

-0.70

 Metatron

-0.69

 Panic

-0.66

×¨

-0.65

ãĥ¤

-0.64

 Primordial

-0.64

 Destroyer

-0.64

POSITIVE LOGITS

eteenth

1.19

iper

1.19

cture

1.17

iors

1.12

ctions

1.09

ipers

1.06

cker

0.93

pei

0.90

ior

0.88

iet

0.86

Activations Density 0.011%

dates, specifically in the format "Jun (day)"

dates and references to specific time periods

No Comments

No Known Activations