INDEX

Explanations

actions or decisions in a structured process

oai_token-act-pair · gpt-3.5-turbo

elements related to roles and functions within structured tasks or processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

 wounding

-0.72

 contrary

-0.69

 misunderstanding

-0.66

 underest

-0.64

 overshadowed

-0.63

Removed

-0.63

inventoryQuantity

-0.63

 unsupported

-0.62

AKING

-0.62

esity

-0.62

POSITIVE LOGITS

 then

1.30

 submit

1.07

then

1.06

 upload

1.05

 decide

1.04

 THEN

1.04

 arrange

1.02

 assign

1.02

 certify

1.00

 determine

1.00

Activations Density 0.443%

actions or decisions in a structured process

elements related to roles and functions within structured tasks or processes

No Comments

No Known Activations