INDEX

Explanations

expressions of regret and apology

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.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

sez

-0.16

 fancy

-0.15

ais

-0.15

 danger

-0.14

 conc

-0.14

 disaster

-0.14

 Tort

-0.14

hon

-0.14

nuts

-0.13

POSITIVE LOGITS

ä¸įå¾Ĺ

0.19

ãģªãģĮãĤī

0.18

Î¹Î²

0.17

 Ð¿ÑĢÐ¸Ð´ÐµÑĤÑģÑı

0.17

 missed

0.16

kul

0.16

awy

0.15

 cannot

0.15

 circumstance

0.15

æµª

0.15

Activations Density 0.111%

expressions of regret and apology

No Comments

No Known Activations