INDEX

Explanations

indicative phrases and statements that express decision-making or personal opinions

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

eton

-0.17

aos

-0.16

auge

-0.15

Cha

-0.15

Î¶

-0.15

Î³Îº

-0.14

Â¶

-0.14

ascimento

-0.14

plode

-0.14

rzy

-0.14

POSITIVE LOGITS

 instead

0.25

 Instead

0.20

Instead

0.19

instead

0.18

anter

0.15

 Ð²Ð¼ÐµÑģÑĤ

0.14

asco

0.14

çļĦæĺ¯

0.14

 only

0.14

Hack

0.14

Activations Density 0.273%

indicative phrases and statements that express decision-making or personal opinions

No Comments

No Known Activations