INDEX

Explanations

references to wildfires

oai_token-act-pair · gpt-3.5-turbo

terms related to wildfires and their impacts

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

 Phar

-0.83

omo

-0.72

atever

-0.70

berman

-0.70

 Virtue

-0.64

racuse

-0.64

atem

-0.64

metics

-0.62

DonaldTrump

-0.62

rium

-0.61

POSITIVE LOGITS

storms

0.98

 raged

0.95

 wildfire

0.92

 raging

0.91

 erupted

0.88

 wildfires

0.88

 exting

0.87

 erupt

0.84

storm

0.81

 roared

0.79

Activations Density 0.029%

references to wildfires

terms related to wildfires and their impacts

No Comments

No Known Activations