INDEX

Explanations

phrases related to waiting or ongoing actions

oai_token-act-pair · gpt-3.5-turbo

phrases that involve continuous action or state while waiting

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

 Christy

-0.70

 Phelps

-0.59

 grit

-0.59

 ..............

-0.59

stan

-0.59

 urgency

-0.58

 ASAP

-0.58

 Verse

-0.58

 quart

-0.58

venants

-0.58

POSITIVE LOGITS

imp

0.73

olphin

0.70

 Blaster

0.67

amiliar

0.65

uble

0.64

ebus

0.64

angering

0.63

eating

0.62

REM

0.62

utical

0.61

Activations Density 0.112%

phrases related to waiting or ongoing actions

phrases that involve continuous action or state while waiting

No Comments

No Known Activations