INDEX

Explanations

phrases that include casual expressions or informal language, particularly those introducing parenthetical information

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

uala

-0.16

-0.14

 Hurt

-0.13

td

-0.13

ouz

-0.13

igit

-0.13

 Unsure

-0.13

wp

-0.13

row

-0.13

POSITIVE LOGITS

krom

0.16

OLON

0.16

à¸Ĳ

0.15

azen

0.15

Ã¡li

0.15

abra

0.14

olon

0.14

SID

0.14

heimer

0.14

 encount

0.14

Activations Density 0.069%

phrases that include casual expressions or informal language, particularly those introducing parenthetical information

No Comments

No Known Activations