INDEX

Explanations

phrases that describe challenges and difficulties in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

xae

-0.16

untime

-0.15

lene

-0.15

itur

-0.15

edom

-0.15

rego

-0.15

antal

-0.14

 likely

-0.14

£

-0.14

uren

-0.14

POSITIVE LOGITS

raki

0.19

 depending

0.17

 sometimes

0.17

uhn

0.17

çĶļèĩ³

0.16

acht

0.15

eras

0.15

opus

0.15

:error

0.15

OMET

0.14

Activations Density 0.186%

phrases that describe challenges and difficulties in various contexts

No Comments

No Known Activations