INDEX

Explanations

phrases expressing a judgment about what is best for a group or individual

oai_token-act-pair · gpt-3.5-turbo

phrases related to the concept of "best interests."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

punk

-0.74

bour

-0.73

qus

-0.71

native

-0.69

rose

-0.64

problem

-0.64

ById

-0.64

versions

-0.63

Obj

-0.63

stellar

-0.62

POSITIVE LOGITS

 electors

0.86

 policymakers

0.77

 Barron

0.70

 Goddard

0.70

ĺħ

0.69

 stewards

0.67

 lett

0.66

 reuse

0.66

 taxpayers

0.63

ually

0.63

Activations Density 0.146%

phrases expressing a judgment about what is best for a group or individual

phrases related to the concept of "best interests."

No Comments

No Known Activations