INDEX

Explanations

proper nouns and high-frequency words indicating significant entities or concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

ovie

-0.16

-0.15

pane

-0.15

à¸¢à¸°

-0.15

brig

-0.14

onas

-0.14

ronym

-0.14

lags

-0.14

 diplom

-0.14

adesh

-0.14

POSITIVE LOGITS

uilder

0.15

zos

0.15

modo

0.15

ifiers

0.15

Ãłng

0.14

bid

0.14

oble

0.14

 Abed

0.14

 Saud

0.14

bett

0.14

Activations Density 0.003%

proper nouns and high-frequency words indicating significant entities or concepts

No Comments

No Known Activations