INDEX

Explanations

words related to numerical or statistical data

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

äºĭåĭĻ

-0.17

zing

-0.16

bol

-0.15

rost

-0.15

FromClass

-0.15

ãĤ

-0.14

 Ø§ÙĨØ¯

-0.14

AML

-0.14

_WR

-0.14

 unst

-0.14

POSITIVE LOGITS

arov

0.17

fur

0.16

 Action

0.16

 supply

0.15

eg

0.15

eln

0.15

ekk

0.15

ern

0.15

Action

0.15

-action

0.15

Activations Density 0.015%

words related to numerical or statistical data

No Comments

No Known Activations