INDEX

Explanations

parentheses and their usage in contexts

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

alion

-0.18

 Dixon

-0.17

ë¥ĺ

-0.14

onte

-0.14

cmc

-0.14

ces

-0.14

<<-

-0.14

nees

-0.13

ation

-0.13

 discard

-0.13

POSITIVE LOGITS

,eg

0.16

 Bloss

0.15

 such

0.14

.plugins

0.13

Tu

0.13

PCP

0.13

tu

0.13

 either

0.13

iris

0.13

 Budd

0.13

Activations Density 0.098%

parentheses and their usage in contexts

No Comments

No Known Activations