INDEX

Explanations

proper nouns related to specific people or brand names

oai_token-act-pair · gpt-3.5-turbo

names and references to specific individuals and entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

ãĤ¡

-0.73

ccording

-0.73

illac

-0.73

utical

-0.72

 laun

-0.72

iability

-0.71

ptives

-0.69

estamp

-0.68

cery

-0.68

 ende

-0.67

POSITIVE LOGITS

Orb

0.87

asar

0.75

 Weaver

0.74

Ops

0.74

urb

0.74

Ack

0.73

itton

0.73

ops

0.72

 Whedon

0.70

Kov

0.69

Activations Density 0.011%

proper nouns related to specific people or brand names

names and references to specific individuals and entities

No Comments

No Known Activations