INDEX

Explanations

instances of chaotic situations or environments

oai_token-act-pair · gpt-3.5-turbo

references to chaos and disorder

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

Footnote

-0.81

 attaches

-0.74

Â®

-0.73

ospons

-0.72

ometer

-0.71

lator

-0.67

 Measure

-0.66

Prof

-0.66

 Citation

-0.66

hook

-0.65

POSITIVE LOGITS

 chaos

3.65

 mayhem

2.32

 chaotic

2.25

 anarchy

2.14

 Chaos

2.10

 pandemonium

2.10

 turmoil

1.87

 havoc

1.78

 carnage

1.75

 tumult

1.67

Activations Density 0.027%

instances of chaotic situations or environments

references to chaos and disorder

No Comments

No Known Activations