INDEX

Explanations

conclusive terms indicating summaries or final thoughts in a text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 1-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_1/width_16k/average_l0_102

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.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

 also

-0.92

 never

-0.73

now

-0.71

 sometimes

-0.68

 still

-0.67

 always

-0.61

 finally

-0.55

 probably

-0.54

 ever

-0.53

see

-0.52

POSITIVE LOGITS

 Moreover

1.49

 However

1.47

 Furthermore

1.46

Furthermore

1.45

Moreover

1.44

 Nevertheless

1.43

 Therefore

1.42

 Additionally

1.41

Additionally

1.41

 Accordingly

1.41

Activations Density 0.265%

conclusive terms indicating summaries or final thoughts in a text

No Comments

No Known Activations