INDEX

Explanations

words ending in "-ifying"

oai_token-act-pair · gpt-3.5-turbo

words that describe processes of clarification or validation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 rides

-0.60

 feat

-0.60

 thor

-0.59

bie

-0.58

dra

-0.57

 Charter

-0.57

adv

-0.55

 Sutton

-0.55

 Sierra

-0.54

 venture

-0.53

POSITIVE LOGITS

ifying

4.32

ifies

3.19

ification

3.08

ified

3.01

ify

2.95

ifiers

2.90

ifier

2.86

ifications

2.68

ifiable

2.17

IFIED

2.11

Activations Density 0.009%

words ending in "-ifying"

words that describe processes of clarification or validation

No Comments

No Known Activations