INDEX

Explanations

repeated phrases indicating conclusions or assertions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.15.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

mybatisplus

-1.07

leſs

-0.76

 Wikimedijinoj

-0.74

Enllaces

-0.73

 nakalista

-0.73

jooq

-0.72

 getPlayer

-0.72

 Pergamon

-0.72

ServiceTest

-0.71

 Dimit

-0.71

POSITIVE LOGITS

the

0.91

 there

0.63

 when

0.61

 although

0.61

由于

0.60

 Although

0.60

if

0.59

由於

0.59

 this

0.58

Although

0.57

Activations Density 0.400%

repeated phrases indicating conclusions or assertions

No Comments

No Known Activations