INDEX

Explanations

phrases related to causality, specifically indicating a consequential result

oai_token-act-pair · gpt-3.5-turbo

phrases indicating causal relationships or consequences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

 lineback

-0.69

 Straw

-0.69

vae

-0.67

ockets

-0.67

arag

-0.65

pent

-0.65

 Offense

-0.64

enta

-0.64

 Mariners

-0.63

 Rusty

-0.62

POSITIVE LOGITS

ainer

0.84

gha

0.73

ãĤ¯

0.72

 thereof

0.71

uary

0.68

ãĥł

0.68

Reviewer

0.66

auder

0.64

iment

0.64

ebin

0.63

Activations Density 0.017%

phrases related to causality, specifically indicating a consequential result

phrases indicating causal relationships or consequences

No Comments

No Known Activations