INDEX

Explanations

keywords related to technical specifications and features of products or services

oai_token-act-pair · gpt-3.5-turbo

terms related to formats and classifications in a technical or administrative context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

yond

-0.93

general

-0.79

olson

-0.76

rift

-0.75

izabeth

-0.74

bishop

-0.73

donald

-0.73

democratic

-0.72

penter

-0.72

otte

-0.72

POSITIVE LOGITS

 Disorder

1.64

 Ratio

1.59

 Units

1.53

 Reduction

1.51

 Facility

1.48

 Program

1.47

 Therapy

1.46

 Calculator

1.45

 Syndrome

1.45

 Awareness

1.44

Activations Density 0.423%

keywords related to technical specifications and features of products or services

terms related to formats and classifications in a technical or administrative context

No Comments

No Known Activations