INDEX

Explanations

phrases indicating conditionality or dependence on circumstances

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

quantities or characteristics that vary based on different circumstances or conditions

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

è£ħ

-0.84

çīĪ

-0.83

İĭ

-0.79

sterdam

-0.78

roo

-0.78

ãĥį

-0.76

ĸļ

-0.76

ļéĨĴ

-0.76

without

-0.75

DragonMagazine

-0.74

POSITIVE LOGITS

 severity

1.01

 circumstance

0.93

 circumstances

0.84

 interpretation

0.82

 factors

0.79

 disposition

0.77

 interpretations

0.76

 quality

0.75

 thresholds

0.74

 calib

0.74

Activations Density 0.170%

phrases indicating conditionality or dependence on circumstances

quantities or characteristics that vary based on different circumstances or conditions

No Comments

No Known Activations