INDEX

Explanations

punctuation marks, particularly periods

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

favicon

-0.16

icot

-0.15

ampler

-0.14

 favicon

-0.14

ificio

-0.14

.sponge

-0.14

orda

-0.14

berman

-0.14

 Metodo

-0.14

ramid

-0.14

POSITIVE LOGITS

 pants

0.16

edback

0.15

EDA

0.14

scan

0.14

ESH

0.13

alous

0.13

opr

0.13

 breakfast

0.13

IX

0.13

 panties

0.13

Activations Density 0.000%

punctuation marks, particularly periods

No Comments

No Known Activations