INDEX

Explanations

mentions of the word "Let" or related variations like "Letting" or "Letters."

oai_token-act-pair · gpt-3.5-turbo

the word "let" in various forms and contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

 holiest

-0.69

DAY

-0.67

BLE

-0.67

liest

-0.66

 cumbers

-0.61

cled

-0.60

PLE

-0.59

ILY

-0.59

foreseen

-0.58

 swipe

-0.57

POSITIVE LOGITS

tered

1.34

ters

1.13

tering

1.13

itia

1.10

ting

1.07

icia

1.02

rons

0.91

rack

0.89

riad

0.89

rans

0.88

Activations Density 0.030%

mentions of the word "Let" or related variations like "Letting" or "Letters."

the word "let" in various forms and contexts

No Comments

No Known Activations