INDEX

Explanations

keywords related to different types of exposures to substances or environments

oai_token-act-pair · gpt-3.5-turbo

references to environmental or chemical exposure

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

otle

-0.72

reme

-0.72

rolley

-0.68

aws

-0.67

 revival

-0.67

lin

-0.64

rolog

-0.64

eteenth

-0.64

seller

-0.64

baum

-0.63

POSITIVE LOGITS

 Exposure

0.92

ibilities

0.90

ibly

0.75

ted

0.74

 exposure

0.73

ibility

0.72

 hazards

0.72

 exposures

0.69

Ð²

0.69

 thereto

0.68

Activations Density 0.035%

keywords related to different types of exposures to substances or environments

references to environmental or chemical exposure

No Comments

No Known Activations