INDEX

Explanations

specific phrases or terms indicating relationships between subjects and their roles in historical or social contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

ettes

-0.16

th

-0.15

agy

-0.15

ierung

-0.15

Ð±Ð°Ñĩ

-0.14

loub

-0.13

 æĪ

-0.13

ÑĢÐ¾Ð²Ð°Ð½Ð¸Ðµ

-0.13

LET

-0.13

ä¸įäºĨ

-0.12

POSITIVE LOGITS

wise

0.18

 nhiÃªn

0.17

unque

0.17

fois

0.17

ìłĢ

0.17

WISE

0.16

Ð¶Ðµ

0.16

quam

0.16

lieÃŁlich

0.15

uiten

0.15

Activations Density 0.143%

specific phrases or terms indicating relationships between subjects and their roles in historical or social contexts

No Comments

No Known Activations