INDEX

Explanations

names starting with "Len"

oai_token-act-pair · gpt-3.5-turbo

mentions of the name "Len" followed by an associated number or identifier

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 contra

-0.91

tc

-0.71

adian

-0.65

eous

-0.65

raised

-0.65

²¾

-0.64

coming

-0.62

straight

-0.62

TF

-0.62

cake

-0.62

POSITIVE LOGITS

Len

4.03

Len

2.61

 Lenn

1.76

len

1.40

 Bren

1.25

len

1.17

 Leonard

1.15

Kap

1.08

 Koen

1.07

 Patty

1.06

Activations Density 0.023%

names starting with "Len"

mentions of the name "Len" followed by an associated number or identifier

No Comments

No Known Activations