INDEX

Explanations

years mentioned in contexts of projections or transformations

oai_token-act-pair · gpt-3.5-turbo

references to specific timeframes or future projections

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

ational

-0.70

ertain

-0.67

inea

-0.65

itto

-0.64

bian

-0.64

ivia

-0.62

 "$:/

-0.62

eta

-0.61

 pains

-0.60

amera

-0.60

POSITIVE LOGITS

 virtue

1.35

products

0.97

laws

0.97

gone

0.85

0.82

catch

0.79

 proxy

0.79

 default

0.79

product

0.78

 means

0.75

Activations Density 0.145%

years mentioned in contexts of projections or transformations

references to specific timeframes or future projections

No Comments

No Known Activations