INDEX

Explanations

comparative terms related to options or choices

oai_token-act-pair · gpt-3.5-turbo

topics related to decision-making or choices

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

igious

-0.71

 Secrets

-0.69

 Visitors

-0.68

 Saud

-0.67

 Intellectual

-0.66

 Brow

-0.66

Ow

-0.64

 Reporting

-0.63

ysis

-0.63

 Lies

-0.62

POSITIVE LOGITS

 preferable

1.54

 preferred

1.24

 feasible

1.23

 advisable

1.17

 viable

1.16

 impractical

1.08

 desirable

1.06

 recommended

1.04

 safer

1.02

 outper

1.02

Activations Density 0.988%

comparative terms related to options or choices

topics related to decision-making or choices

No Comments

No Known Activations