INDEX

Explanations

terms related to eligibility or meeting specific criteria

oai_token-act-pair · gpt-3.5-turbo

the term "qualify" in various contexts related to eligibility and criteria

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

RH

-0.65

 Alps

-0.62

bru

-0.60

 revolving

-0.59

 Kiss

-0.59

 Lans

-0.58

 goodbye

-0.58

axis

-0.58

Swe

-0.57

 Gardens

-0.57

POSITIVE LOGITS

 qualify

1.23

 qualified

0.98

ifications

0.94

 qualifies

0.94

 qualifications

0.89

rieved

0.88

ifies

0.88

ifiers

0.83

dinand

0.82

 disqualified

0.82

Activations Density 0.005%

terms related to eligibility or meeting specific criteria

the term "qualify" in various contexts related to eligibility and criteria

No Comments

No Known Activations