INDEX

Explanations

timestamps and numerical values associated with time

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

iki

-0.18

-0.15

endo

-0.15

odo

-0.14

hab

-0.14

 become

-0.14

flow

-0.14

POSITIVE LOGITS

ERRU

0.18

icina

0.15

 rodin

0.15

adÄĽ

0.15

Ð°ÑĦ

0.14

ptron

0.14

_pm

0.14

Ã¢u

0.14

-prefix

0.14

Æ¡n

0.14

Activations Density 0.026%

timestamps and numerical values associated with time

No Comments

No Known Activations