INDEX

Explanations

names and identifiers of people involved in a document

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

isto

-0.18

Foo

-0.17

 Cald

-0.15

iÅ¾

-0.15

eme

-0.15

ulen

-0.14

ardon

-0.14

izzard

-0.14

prit

-0.14

abay

-0.14

POSITIVE LOGITS

pons

0.17

lamaz

0.15

 nedeni

0.14

ÅĻÃŃz

0.14

 optical

0.14

 Schul

0.14

dll

0.13

Î³Îº

0.13

ourn

0.13

Kun

0.13

Activations Density 0.015%

names and identifiers of people involved in a document

No Comments

No Known Activations