INDEX

Explanations

references to time and durations related to events or situations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

either

-0.17

ccount

-0.16

 either

-0.16

 EITHER

-0.16

ifest

-0.15

ÑĭÑĪ

-0.14

 rast

-0.14

ochen

-0.14

ternet

-0.14

ainless

-0.13

POSITIVE LOGITS

 overall

0.17

illez

0.14

!important

0.14

lernen

0.14

 Overall

0.14

 generations

0.14

endir

0.14

 BÃ¶l

0.13

/browse

0.13

 thousands

0.13

Activations Density 0.023%

references to time and durations related to events or situations

No Comments

No Known Activations