INDEX

Explanations

multi-word phrases related to rankings or positions

oai_token-act-pair · gpt-3.5-turbo

references to numerical rankings or positions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

rador

-0.82

outer

-0.73

Materials

-0.69

 Shroud

-0.66

 Kafka

-0.64

rait

-0.62

raught

-0.62

heid

-0.62

axter

-0.61

hire

-0.61

POSITIVE LOGITS

plates

1.01

plate

0.87

enance

0.85

esses

0.82

 crunch

0.79

metry

0.77

number

0.74

atur

0.71

eah

0.70

ãĥ¼ãĥ³

0.69

Activations Density 0.046%

multi-word phrases related to rankings or positions

references to numerical rankings or positions

No Comments

No Known Activations