INDEX

Explanations

phrases indicating the first instance or achievement in a given context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

åħ¥ãĤĬ

-0.14

atto

-0.14

ÑīÐµ

-0.14

-0.13

rve

-0.13

lico

-0.13

usz

-0.13

roj

-0.13

uchos

-0.13

POSITIVE LOGITS

SEA

0.16

memberof

0.15

-ever

0.14

 ties

0.14

adows

0.14

å®Įæķ´

0.14

iasco

0.13

SEQ

0.13

(s

0.13

Activations Density 0.021%

phrases indicating the first instance or achievement in a given context

No Comments

No Known Activations

phrases indicating the first instance or achievement in a given context

No Comments

No Known Activations