INDEX

Explanations

references to specific technical or scientific terms related to classifications or types

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.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

closedir

-0.56

outdir

-0.47

]");

-0.47

theless

-0.45

 Geor

-0.45

stylesheet

-0.44

 ymin

-0.44

[];

-0.44

olid

-0.43

ly

-0.43

POSITIVE LOGITS

TC

1.02

TC

0.98

tc

0.92

tc

0.82

TCU

0.77

Tc

0.75

Tc

0.71

TCA

0.71

TCA

0.68

MTC

0.65

Activations Density 0.017%

references to specific technical or scientific terms related to classifications or types

No Comments

No Known Activations