INDEX

Explanations

references to specific geographical or institutional communities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 15-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L15R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.15.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

Å¥

-0.17

ersist

-0.16

 Consort

-0.16

adla

-0.14

WEBPACK

-0.14

 háºŃu

-0.13

hana

-0.13

arium

-0.13

 hers

-0.13

Ð¾Ð·

-0.13

POSITIVE LOGITS

who

0.24

 whom

0.21

who

0.20

qui

0.16

sn

0.14

ÙħØ§ÙĨÛĮ

0.14

 fame

0.14

ornings

0.14

istent

0.14

Who

0.13

Activations Density 0.152%

references to specific geographical or institutional communities

No Comments

No Known Activations