INDEX

Explanations

mentions of the word 'drop' with varying intensities

oai_token-act-pair · gpt-3.5-turbo

variations of the word "dro" in different contexts

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Clause

-0.75

ãĥĨãĤ£

-0.74

Ou

-0.72

 Origins

-0.72

 reconciliation

-0.72

Myth

-0.71

 Trident

-0.68

ãĥ¯ãĥ³

-0.65

geist

-0.65

é¾

-0.65

POSITIVE LOGITS

dro

1.25

plets

1.16

plet

1.02

ppers

0.96

dro

0.90

pper

0.88

ble

0.86

umph

0.84

pless

0.83

Dro

0.78

Activations Density 0.010%

mentions of the word 'drop' with varying intensities

variations of the word "dro" in different contexts

No Comments

No Known Activations