INDEX

Explanations

instances of the word "ironically" and similar variations, indicating a focus on irony and unexpected situations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 19-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L19R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.19.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

Ø±Ø¬

-0.17

isse

-0.16

endale

-0.15

ÏĥÎ¹Î¿

-0.14

 Ð¸Ð¼Ñĥ

-0.14

 ÑģÐ¾Ð¼

-0.14

Ð»Ð¸ÑĨ

-0.14

ìķ½

-0.14

sao

-0.14

åĨµ

-0.14

POSITIVE LOGITS

 enough

0.41

 Enough

0.27

 sufficient

0.20

Enough

0.18

ingly

0.17

åł´

0.16

ë°Ķ

0.16

(?

0.15

aller

0.15

å¤Ł

0.15

Activations Density 0.026%

instances of the word "ironically" and similar variations, indicating a focus on irony and unexpected situations

No Comments

No Known Activations