INDEX

Explanations

instances of reported speech, specifically phrases that indicate someone is conveying information

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

zet

-0.16

 Springs

-0.15

 Featured

-0.15

oud

-0.15

zes

-0.14

Toy

-0.14

 index

-0.14

rec

-0.14

ic

-0.14

 indexes

-0.14

POSITIVE LOGITS

yb

0.16

chwitz

0.16

Scalars

0.15

Vys

0.15

ProcessEvent

0.15

 olmam

0.14

gba

0.14

quences

0.14

verts

0.14

">ÃĹ</

0.14

Activations Density 0.023%

instances of reported speech, specifically phrases that indicate someone is conveying information

No Comments

No Known Activations