INDEX

Explanations

numbers related to quantities or measurements

oai_token-act-pair · gpt-3.5-turbo

numerical values, particularly those related to counts or measurements

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

bara

-0.92

oru

-0.90

lace

-0.76

iguous

-0.70

folk

-0.69

ndra

-0.66

gart

-0.65

pared

-0.65

Ã§a

-0.65

lay

-0.65

POSITIVE LOGITS

 ILCS

1.25

0.75

 chars

0.74

 "$:/

0.72

dB

0.71

0.70

dB

0.68

 sshd

0.66

 Travels

0.66

EMBER

0.66

Activations Density 0.021%

numbers related to quantities or measurements

numerical values, particularly those related to counts or measurements

No Comments

No Known Activations