INDEX

Explanations

locations and geographical names in the text

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

Å¥

-0.14

nid

-0.14

 æ¢

-0.14

rej

-0.14

nock

-0.14

qid

-0.13

_FAST

-0.13

sep

-0.13

POSITIVE LOGITS

ajan

0.17

oire

0.17

ogne

0.15

iner

0.14

amps

0.14

çĵľ

0.14

engo

0.14

achts

0.14

ĳ

0.14

icz

0.14

Activations Density 0.067%

locations and geographical names in the text

No Comments

No Known Activations