INDEX

Explanations

references to the name "Richard" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

EF

-0.16

ural

-0.15

amer

-0.15

agher

-0.15

bjerg

-0.15

PCA

-0.15

å¥´

-0.15

arend

-0.15

itical

-0.15

POSITIVE LOGITS

sons

0.26

son

0.25

SON

0.22

ÑģÐ¾Ð½

0.21

 Nixon

0.18

sonian

0.17

ÏĥÎ¿Î½

0.17

lineno

0.17

hof

0.15

loggedin

0.15

Activations Density 0.012%

references to the name "Richard" in various contexts

No Comments

No Known Activations