INDEX

Explanations

the word "den" in various contexts

oai_token-act-pair · gpt-3.5-turbo

the word "den" and its variations in context

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

ramid

-0.73

lda

-0.70

 Carbuncle

-0.70

drive

-0.69

NetMessage

-0.69

rocal

-0.66

toggle

-0.65

 Mandatory

-0.65

raltar

-0.62

zzo

-0.62

POSITIVE LOGITS

den

1.35

izens

1.22

unci

1.14

unciation

1.07

ãĤ§

0.94

igration

0.92

ounces

0.88

izen

0.87

ãĥķãĤ¡

0.84

igr

0.83

Activations Density 0.006%

the word "den" in various contexts

the word "den" and its variations in context

No Comments

No Known Activations