INDEX

Explanations

phrases related to reports and official statements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

 Pont

-0.16

 Verm

-0.15

 titular

-0.15

 perish

-0.15

 pleasing

-0.14

isman

-0.14

 Sadd

-0.14

Ã©ri

-0.14

ave

-0.13

POSITIVE LOGITS

stants

0.16

ç¿Ķ

0.14

ICON

0.14

ulan

0.14

ÏĩÎ·

0.14

resentation

0.14

ãĥ³ãĥĩãĤ£

0.14

 xlink

0.13

conds

0.13

ENU

0.13

Activations Density 0.036%

phrases related to reports and official statements

No Comments

No Known Activations