INDEX

Explanations

references to the word "Wolf" and its variations in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

ment

-0.19

nga

-0.16

Ð¸ÑĪ

-0.14

ãĥ¡ãĥ³ãĥĪ

-0.14

town

-0.14

åłĤ

-0.14

gado

-0.14

åĪ·

-0.14

alm

-0.14

AMENT

-0.13

POSITIVE LOGITS

lets

0.17

Ø¹Ùħ

0.15

letal

0.15

cest

0.15

endale

0.15

çĪµ

0.15

illard

0.15

à¹Ģà¸§à¸Ńà¸£

0.14

LOT

0.14

ä»®

0.14

Activations Density 0.027%

references to the word "Wolf" and its variations in the text

No Comments

No Known Activations