INDEX

Explanations

instances of phrases related to indications or signals

oai_token-act-pair · gpt-3.5-turbo

phrases indicating the presence of signs or indications of various situations or conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 Cosponsors

-0.86

uristic

-0.75

apest

-0.74

estyles

-0.71

ades

-0.71

taboola

-0.70

Merit

-0.70

pps

-0.69

alez

-0.69

terms

-0.68

POSITIVE LOGITS

 wrongdoing

1.22

 impending

1.16

 tampering

1.13

 weakness

1.05

 innocence

1.05

 desperation

1.05

 collusion

1.04

 criminality

1.02

 instability

1.01

 improvement

1.01

Activations Density 0.140%

instances of phrases related to indications or signals

phrases indicating the presence of signs or indications of various situations or conditions

No Comments

No Known Activations