INDEX

Explanations

words that convey proximity or closeness

oai_token-act-pair · gpt-3.5-turbo

references to proximity or nearness

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

mA

-0.82

AIN

-0.80

uria

-0.77

NZ

-0.76

ICAN

-0.74

heim

-0.72

ãĤ¸

-0.70

ican

-0.70

alian

-0.68

mal

-0.68

POSITIVE LOGITS

 proximity

1.02

sighted

0.99

 minded

0.90

 resemblance

0.79

 confinement

0.79

 enough

0.76

 vicinity

0.75

 thereto

0.74

 neighbours

0.71

 clos

0.71

Activations Density 0.023%

words that convey proximity or closeness

references to proximity or nearness

No Comments

No Known Activations