INDEX

Explanations

adverbs describing actions in a positive light

oai_token-act-pair · gpt-3.5-turbo

adverbs that express emotional states or attitudes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 disproportion

-0.79

Liter

-0.72

VL

-0.71

senal

-0.70

 purely

-0.66

LER

-0.65

amy

-0.63

RAW

-0.62

quickShipAvailable

-0.62

rification

-0.62

POSITIVE LOGITS

 embraced

0.95

 awaited

0.94

 complied

0.91

 apologised

0.91

 conceded

0.88

 apologized

0.88

 transitioned

0.88

 dispatched

0.86

 accepted

0.86

 parted

0.85

Activations Density 0.105%

adverbs describing actions in a positive light

adverbs that express emotional states or attitudes

No Comments

No Known Activations