INDEX

Explanations

words related to discipline or punishment

oai_token-act-pair · gpt-3.5-turbo

instances of the substring "pl"

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

 Coffin

-0.69

 Jarrett

-0.67

 Aber

-0.66

 Elias

-0.66

 Curse

-0.65

 lockout

-0.62

 bonds

-0.61

 adversary

-0.61

Rah

-0.61

 Emir

-0.60

POSITIVE LOGITS

pl

4.40

pler

1.84

plain

1.73

plant

1.67

PL

1.63

plan

1.61

ple

1.59

plin

1.53

pled

1.50

Pl

1.50

Activations Density 0.005%

words related to discipline or punishment

instances of the substring "pl"

No Comments

No Known Activations