INDEX

Explanations

references to the second person "you" 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

urus

-0.15

ÐµÐ¿ÑĤÐ¸

-0.15

ÏĦÏģÎ¿

-0.15

agos

-0.15

åħ¥ãĤĮ

-0.15

á»©c

-0.15

CHA

-0.14

kla

-0.14

lest

-0.14

igest

-0.14

POSITIVE LOGITS

 debt

0.19

 should

0.17

 definitely

0.17

 Debt

0.17

ãģĤãĤĬãģĮãģ¨ãģĨ

0.15

 might

0.15

 better

0.15

 hÃ£y

0.15

 definite

0.15

Activations Density 0.071%

references to the second person "you" in various contexts

No Comments

No Known Activations