INDEX

Explanations

expressions of enjoyment and positive experiences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

opher

-0.16

 Wake

-0.14

angers

-0.14

onders

-0.14

eyer

-0.14

 vá»įng

-0.13

 Concept

-0.13

æ¤į

-0.13

ÑģÑĤÐ¸

-0.13

utherford

-0.13

POSITIVE LOGITS

ayne

0.17

 ---------------------------------------------------------------------------↵

0.15

 experience

0.14

æĢ§çļĦ

0.14

{{{

0.14

edula

0.14

 experiences

0.14

idon

0.13

 ÑĩÐ°

0.13

fte

0.13

Activations Density 0.097%

expressions of enjoyment and positive experiences

No Comments

No Known Activations