INDEX

Explanations

phrases indicating completion or past actions

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "been" and its variations, indicating past actions or states

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

Downloadha

-0.77

quickShipAvailable

-0.73

CST

-0.69

morning

-0.65

Ranked

-0.62

especially

-0.60

kai

-0.60

etheless

-0.60

Pacific

-0.60

awaru

-0.59

POSITIVE LOGITS

gebra

0.74

ledged

0.73

ential

0.72

entials

0.71

usercontent

0.67

 acknow

0.66

ailable

0.65

izable

0.64

nance

0.64

ancial

0.63

Activations Density 0.173%

phrases indicating completion or past actions

occurrences of the word "been" and its variations, indicating past actions or states

No Comments

No Known Activations