INDEX

Explanations

phrases related to gaining or losing momentum

oai_token-act-pair · gpt-3.5-turbo

references to the concept of momentum in various contexts

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

tein

-0.82

mits

-0.73

ramid

-0.73

mis

-0.72

bery

-0.71

haps

-0.71

missions

-0.71

pper

-0.69

Interstitial

-0.68

sa

-0.68

POSITIVE LOGITS

 momentum

1.43

wcs

0.90

 exha

0.89

 inertia

0.84

 impetus

0.83

 pend

0.77

ernaut

0.76

 withd

0.74

inventoryQuantity

0.73

 cohesion

0.73

Activations Density 0.008%

phrases related to gaining or losing momentum

references to the concept of momentum in various contexts

No Comments

No Known Activations