INDEX

Explanations

phrases related to making a task more difficult or easier

oai_token-act-pair · gpt-3.5-turbo

terms related to processes, conditions, or concepts that involve significant change or impact

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

-0.82

 Entered

-0.75

cair

-0.67

Shares

-0.63

utm

-0.62

idth

-0.62

edIn

-0.62

 Cosponsors

-0.61

 greets

-0.61

 Sins

-0.61

POSITIVE LOGITS

 obsolete

1.12

 easier

1.07

 safer

1.05

 redundant

1.00

 unavailable

1.00

 impossible

0.99

 inaccessible

0.98

 accessible

0.98

 worthwhile

0.95

 irrelevant

0.94

Activations Density 0.202%

phrases related to making a task more difficult or easier

terms related to processes, conditions, or concepts that involve significant change or impact

No Comments

No Known Activations