INDEX

Explanations

capitalized names and references to locations or significant terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

vana

-0.17

ioni

-0.17

.gdx

-0.16

 Hague

-0.15

Ð¾ÐºÐ¸

-0.15

olie

-0.15

uv

-0.15

resse

-0.14

ists

-0.14

tera

-0.14

POSITIVE LOGITS

DIG

0.19

Dig

0.17

DIG

0.17

Sa

0.17

dig

0.17

dig

0.16

Dig

0.16

Digital

0.16

Digit

0.15

 Degree

0.15

Activations Density 0.036%

capitalized names and references to locations or significant terms

No Comments

No Known Activations