INDEX

Explanations

mentions of the word "who" 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 @ 3-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L3R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.3.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

ardown

-0.15

illes

-0.15

yen

-0.15

å¼ķãģį

-0.14

gaard

-0.14

å¹ķ

-0.13

Î¹Ïĥ

-0.13

ÑĢÐ°Ð·Ð´

-0.13

èĦļ

-0.13

SURE

-0.13

POSITIVE LOGITS

oping

0.17

upon

0.16

akin

0.15

endale

0.15

tridge

0.14

akes

0.14

osh

0.14

annel

0.14

Ð²Ð¾ÑĢ

0.13

 sinc

0.13

Activations Density 0.122%

mentions of the word "who" in various contexts

No Comments

No Known Activations