INDEX

Explanations

medical and health-related terms

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate attention, awareness, or desire in conversations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

 Quentin

-0.66

 drafts

-0.58

 Sloan

-0.58

 DMCA

-0.57

 Sched

-0.56

 Metatron

-0.55

 Latter

-0.55

Fed

-0.54

BLM

-0.54

 casualty

-0.54

POSITIVE LOGITS

ÙĴ

0.74

ãĢĤ

0.67

enium

0.66

.:

0.66

uggle

0.65

_-_

0.65

ummies

0.65

çĲ

0.65

ortmund

0.64

 anyways

0.64

Activations Density 0.387%

medical and health-related terms

phrases that indicate attention, awareness, or desire in conversations

No Comments

No Known Activations