INDEX

Explanations

instances of the word "too" indicating an excess or limitation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

*/;

-0.81

ViewFeatures

-0.81

roves

-0.76

)");

-0.76

APIView

-0.74

 sement

-0.74

{{$

-0.73

Anas

-0.73

Kennedy

-0.73

}")

-0.73

POSITIVE LOGITS

Too

2.02

too

2.00

Too

1.92

too

1.86

TOO

1.84

TOO

1.73

troppo

1.23

 zbyt

1.22

 слишком

1.17

 terlalu

1.16

Activations Density 0.058%

instances of the word "too" indicating an excess or limitation

No Comments

No Known Activations