INDEX

Explanations

phrases indicating recognition or acceptance of a fact or situation

oai_token-act-pair · gpt-3.5-turbo

instances of the word "acknowledge" in various forms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

inj

-0.68

chn

-0.68

ibur

-0.63

carb

-0.63

bin

-0.63

 Hungry

-0.63

cycl

-0.62

hire

-0.62

bs

-0.61

rou

-0.61

POSITIVE LOGITS

 acknowledges

1.15

 acknowledge

1.10

 acknowledgment

1.00

 acknowledging

0.99

 acknowledged

0.99

ledged

0.94

 acknowled

0.90

 acknow

0.86

 acknowledgement

0.86

 fault

0.77

Activations Density 0.016%

phrases indicating recognition or acceptance of a fact or situation

instances of the word "acknowledge" in various forms

No Comments

No Known Activations