INDEX

Explanations

potential outcomes or possibilities indicated by the word "could."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

%X

-0.16

usercontent

-0.15

etry

-0.15

ement

-0.15

lod

-0.15

.EqualTo

-0.14

paged

-0.14

eting

-0.14

allas

-0.14

ford

-0.14

POSITIVE LOGITS

nt

0.18

 Schul

0.16

 easily

0.15

 potentially

0.15

ABCDEFGHI

0.14

 potential

0.14

ÏĢÎ¿ÏĦÎµ

0.14

ìĶ

0.14

ld

0.13

 subs

0.13

Activations Density 0.067%

potential outcomes or possibilities indicated by the word "could."

No Comments

No Known Activations