INDEX

Explanations

temporal references and time indicators

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 7-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L7R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.7.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

yst

-0.17

bdb

-0.15

WARDED

-0.15

edia

-0.15

 Union

-0.15

aminer

-0.15

 UNION

-0.14

ãĥ¥ãĥ¼

-0.14

Union

-0.14

 vigil

-0.14

POSITIVE LOGITS

 Sesso

0.18

åī¯

0.15

Sharper

0.15

afe

0.15

Ã©ra

0.15

 ruku

0.14

 Lange

0.14

 Ware

0.13

çĦ

0.13

declare

0.13

Activations Density 0.218%

temporal references and time indicators

No Comments

No Known Activations