INDEX

Explanations

names related to 'Al' with varying activations

oai_token-act-pair · gpt-3.5-turbo

the name "Al" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

 Chaser

-0.84

DragonMagazine

-0.82

olicy

-0.81

ãģį

-0.80

lihood

-0.70

intendent

-0.68

 Jagu

-0.67

 Beware

-0.66

 Seah

-0.66

ãĥ¼ãĥĨãĤ£

-0.65

POSITIVE LOGITS

manac

1.25

gebra

1.18

gorithm

1.08

gorith

1.07

aska

1.05

ameda

1.04

umni

1.04

onso

1.04

addin

1.02

ibi

0.98

Activations Density 0.027%

names related to 'Al' with varying activations

the name "Al" in various contexts

No Comments

No Known Activations