INDEX

Explanations

specific mentions of nuclear reactors

oai_token-act-pair · gpt-3.5-turbo

references to nuclear reactors and their associated conditions or incidents

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

 Agric

-0.74

Beg

-0.69

lihood

-0.69

ean

-0.68

ãĤĭ

-0.67

romeda

-0.67

 Merrill

-0.66

 Lives

-0.66

umbnail

-0.66

 Emir

-0.65

POSITIVE LOGITS

 reactor

1.01

 reactors

1.01

ptions

0.79

inery

0.78

 installations

0.78

 dismant

0.75

 towers

0.75

uclear

0.73

emetery

0.73

 centrif

0.72

Activations Density 0.026%

specific mentions of nuclear reactors

references to nuclear reactors and their associated conditions or incidents

No Comments

No Known Activations