INDEX

Explanations

specific nouns or keywords indicating locations, roles, and significant identifiers

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

enas

-0.17

okers

-0.15

iod

-0.15

ang

-0.14

angs

-0.14

ajo

-0.14

oodle

-0.14

jal

-0.13

Äĩe

-0.13

etler

-0.13

POSITIVE LOGITS

gn

0.15

hq

0.15

ges

0.14

istol

0.14

 sandwich

0.14

 Sand

0.13

ÐĲÑĢÑħÑĸÐ²

0.13

vyk

0.13

-Requested

0.13

 Sandwich

0.13

Activations Density 0.003%

specific nouns or keywords indicating locations, roles, and significant identifiers

No Comments

No Known Activations

specific nouns or keywords indicating locations, roles, and significant identifiers

No Comments

No Known Activations