INDEX

Explanations

phrases indicating inability to perform an action or achieve a result

oai_token-act-pair · gpt-3.5-turbo

instances of the word "unable" indicating incapacity or lack of ability

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

soType

-0.73

 Tradition

-0.72

agate

-0.71

oak

-0.68

RAFT

-0.66

afort

-0.65

eg

-0.65

soDeliveryDate

-0.65

gone

-0.64

eye

-0.64

POSITIVE LOGITS

ioned

0.88

 compe

0.88

icably

0.77

 adequately

0.76

poke

0.72

xes

0.71

 necessarily

0.70

bos

0.68

ison

0.67

lessly

0.67

Activations Density 0.024%

phrases indicating inability to perform an action or achieve a result

instances of the word "unable" indicating incapacity or lack of ability

No Comments

No Known Activations