INDEX

Explanations

assertions about personal beliefs and emotional experiences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

ãĤ¤ãĥī

-0.15

iyas

-0.15

ventus

-0.15

NS

-0.14

æĤª

-0.14

Î´Î¬

-0.14

ropolis

-0.14

.aspx

-0.14

oples

-0.13

 Gale

-0.13

POSITIVE LOGITS

 would

0.21

 should

0.20

 could

0.19

 will

0.19

 might

0.18

is

0.17

 ought

0.17

 must

0.16

 SHOULD

0.15

OULD

0.15

Activations Density 0.072%

assertions about personal beliefs and emotional experiences

No Comments

No Known Activations