INDEX

Explanations

references to non-reflective materials or concepts

oai_token-act-pair · gpt-3.5-turbo

the term "non" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 Tycoon

-0.96

 Oaks

-0.67

 Seasons

-0.64

 Takes

-0.62

 downfall

-0.61

 guts

-0.61

 Grill

-0.60

IUM

-0.60

 trove

-0.59

 Learns

-0.59

POSITIVE LOGITS

chal

1.33

linear

1.32

verbal

1.28

stop

1.22

lethal

1.22

fiction

1.21

etheless

1.17

violence

1.17

compliance

1.17

binary

1.15

Activations Density 0.028%

references to non-reflective materials or concepts

the term "non" in various contexts

No Comments

No Known Activations