INDEX

Explanations

terms related to numerical identifiers and ratings

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

âĢ¦↵

-0.17

acula

-0.14

ip

-0.14

Âł

-0.14

 tradition

-0.14

ellen

-0.14

anson

-0.14

 âĢ¦↵

-0.13

op

-0.13

ism

-0.13

POSITIVE LOGITS

elerik

0.15

ledi

0.15

intColor

0.14

ÙĪØ±Ø´

0.14

ãĥ³ãĥĶ

0.14

 addCriterion

0.14

/compiler

0.14

GGLE

0.14

/assert

0.13

ickname

0.13

Activations Density 0.084%

terms related to numerical identifiers and ratings

No Comments

No Known Activations