INDEX

Explanations

references to witnesses giving testimonies or statements

oai_token-act-pair · gpt-3.5-turbo

references to eyewitnesses in a narrative

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

cair

-0.77

atism

-0.75

arm

-0.74

obic

-0.72

ono

-0.71

Saharan

-0.71

chan

-0.70

erm

-0.70

erenn

-0.69

gr

-0.66

POSITIVE LOGITS

 witnesses

1.32

 Witnesses

1.11

 testimonies

1.10

 sightings

1.03

ewitness

0.99

Witness

0.94

 witness

0.92

 Witness

0.86

 Cosponsors

0.84

 eyewitness

0.81

Activations Density 0.011%

references to witnesses giving testimonies or statements

references to eyewitnesses in a narrative

No Comments

No Known Activations