INDEX

Explanations

pronouns related to possession, particularly "his."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

 Steele

-0.16

horizontal

-0.14

åŃĿ

-0.14

eff

-0.14

ault

-0.13

.UserInfo

-0.13

.camel

-0.13

çļĦåľ°æĸ¹

-0.13

Fir

-0.13

ants

-0.13

POSITIVE LOGITS

/her

0.33

own

0.24

panic

0.20

own

0.17

/she

0.17

ABCDEFGHIJKLMNOP

0.16

jeta

0.16

immel

0.15

editary

0.15

gÃ¼n

0.15

Activations Density 0.223%

pronouns related to possession, particularly "his."

No Comments

No Known Activations