INDEX

Explanations

references to atomic particles, particularly focusing on the behavior of atoms in different scenarios

oai_token-act-pair · gpt-3.5-turbo

word repetitions related to atoms and their properties

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

 Fargo

-0.81

PE

-0.76

 Cosponsors

-0.74

Guard

-0.73

VICE

-0.71

 Leone

-0.69

inav

-0.69

 Sparrow

-0.68

HCR

-0.66

 Angeles

-0.65

POSITIVE LOGITS

 atoms

1.20

 atom

1.09

aceous

1.01

ctrl

0.96

ysis

0.84

ohyd

0.84

atom

0.82

oxin

0.82

ucle

0.81

atomic

0.81

Activations Density 0.013%

references to atomic particles, particularly focusing on the behavior of atoms in different scenarios

word repetitions related to atoms and their properties

No Comments

No Known Activations