INDEX

Explanations

instances of shaking or nodding, particularly in relation to emotional responses

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

anÃ¡

-0.18

hone

-0.16

 Ø§ÙĦØ¯Ø±

-0.15

unnable

-0.15

åĭĻ

-0.15

acting

-0.15

Backing

-0.15

agini

-0.14

ÃĹ↵↵

-0.14

awks

-0.14

POSITIVE LOGITS

ingly

0.26

ly

0.16

avian

0.15

Ã¸j

0.15

-fast

0.15

-proof

0.15

ring

0.15

oord

0.15

kel

0.14

htar

0.14

Activations Density 0.049%

instances of shaking or nodding, particularly in relation to emotional responses

No Comments

No Known Activations