INDEX

Explanations

positions or roles held by individuals in various organizations or associations

oai_token-act-pair · gpt-3.5-turbo

phrases indicating roles, positions, or titles related to individuals or organizations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

anooga

-0.71

 stranger

-0.69

 specificity

-0.69

 unbeliev

-0.66

soType

-0.64

dayName

-0.63

 raining

-0.63

lessness

-0.63

 rationale

-0.61

 horr

-0.61

POSITIVE LOGITS

 sorts

0.82

 Lanka

0.68

guard

0.66

ggles

0.66

cycle

0.64

hovah

0.64

 Turtle

0.64

 Plymouth

0.64

adden

0.63

 Hemisphere

0.63

Activations Density 0.230%

positions or roles held by individuals in various organizations or associations

phrases indicating roles, positions, or titles related to individuals or organizations

No Comments

No Known Activations