INDEX

Explanations

future tense forms of verbs

oai_token-act-pair · gpt-3.5-turbo

phrases expressing inevitability or negation regarding future outcomes

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

 antioxid

-0.77

redients

-0.69

ahime

-0.63

 portfolios

-0.63

Leaks

-0.62

 occasionally

-0.62

Newsletter

-0.60

 upgr

-0.58

igl

-0.58

 periodically

-0.55

POSITIVE LOGITS

nor

0.97

before

0.80

theless

0.78

 doubted

0.77

 imagined

0.71

rences

0.70

 penny

0.70

 before

0.69

 dime

0.69

thood

0.67

Activations Density 0.170%

future tense forms of verbs

phrases expressing inevitability or negation regarding future outcomes

No Comments

No Known Activations