INDEX

Explanations

unique identifiers and structural elements in a text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 28-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L28R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.28.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Truthy

-0.14

 till

-0.14

enerima

-0.14

 Till

-0.14

æ¸¡

-0.14

 profit

-0.13

 attempt

-0.13

GPL

-0.13

Spacer

-0.13

 obvious

-0.13

POSITIVE LOGITS

ìļ°ë¦¬

0.22

erd

0.19

aal

0.18

ewe

0.16

 article

0.15

ises

0.15

ale

0.15

Å¡ak

0.15

ignet

0.14

aria

0.14

Activations Density 0.002%

unique identifiers and structural elements in a text

No Comments

No Known Activations