INDEX

Explanations

phrases with the word "trying"

oai_token-act-pair · gpt-3.5-turbo

verbs related to attempts or efforts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

CVE

-0.74

 Upton

-0.74

imov

-0.69

-0.66

inant

-0.65

anas

-0.64

 Debor

-0.64

 Serie

-0.62

 Zamb

-0.62

lance

-0.61

POSITIVE LOGITS

culosis

0.86

phrine

0.82

gged

0.70

gging

0.70

elly

0.68

cks

0.68

lyak

0.67

ctors

0.65

onies

0.65

rums

0.65

Activations Density 0.017%

phrases with the word "trying"

verbs related to attempts or efforts

No Comments

No Known Activations