INDEX

Explanations

references to podcasts and their entertaining qualities

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

ocks

-0.18

ress

-0.15

lexer

-0.15

_MT

-0.15

omer

-0.14

 prop

-0.14

atch

-0.14

iente

-0.14

inen

-0.14

oire

-0.14

POSITIVE LOGITS

.fromFunction

0.16

/rfc

0.16

.glide

0.15

 bÃ³ng

0.14

ibase

0.14

.gdx

0.14

asto

0.14

BOSE

0.14

_stderr

0.14

.strictEqual

0.14

Activations Density 0.022%

references to podcasts and their entertaining qualities

No Comments

No Known Activations