INDEX

Explanations

occurrences of the word "the" and related phrases indicating breaking or changing patterns or norms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

irth

-0.17

uzzy

-0.14

bsd

-0.14

 ÙħØ®Øª

-0.14

 turb

-0.14

bourg

-0.13

erals

-0.13

eterminate

-0.13

rif

-0.13

RIX

-0.13

POSITIVE LOGITS

 barrier

0.31

 barriers

0.29

 mold

0.28

 mould

0.28

 Barrier

0.26

 deadlock

0.26

 seal

0.25

 bonds

0.24

Barrier

0.24

 spell

0.24

Activations Density 0.027%

occurrences of the word "the" and related phrases indicating breaking or changing patterns or norms

No Comments

No Known Activations