INDEX

Explanations

phrases related to legal actions or accusations

oai_token-act-pair · gpt-3.5-turbo

the occurrences of a specific name or term related to important individuals or entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

Ath

-0.65

 critical

-0.63

exc

-0.61

 staged

-0.60

sub

-0.60

 catalyst

-0.59

bul

-0.58

 phased

-0.58

gut

-0.58

Nan

-0.57

POSITIVE LOGITS

ords

5.00

ord

2.31

ording

2.14

ORD

1.98

irms

1.29

orders

1.22

orde

1.19

ridges

1.13

efficients

1.10

order

1.05

Activations Density 0.013%

phrases related to legal actions or accusations

the occurrences of a specific name or term related to important individuals or entities

No Comments

No Known Activations