INDEX

Explanations

phrases related to geographical or environmental descriptions

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

anker

-0.15

 Cliff

-0.14

 quantitative

-0.14

 Closed

-0.14

chez

-0.13

-oper

-0.13

atten

-0.13

 chod

-0.13

atel

-0.13

ehen

-0.13

POSITIVE LOGITS

 lying

0.29

lie

0.25

 lies

0.24

Stretch

0.23

lies

0.23

 astr

0.23

 border

0.21

 Stretch

0.20

stretch

0.20

 stretching

0.20

Activations Density 0.175%

phrases related to geographical or environmental descriptions

No Comments

No Known Activations