INDEX

Explanations

references to various forms of training

oai_token-act-pair · gpt-3.5-turbo

references to training programs or sessions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

gets

-0.66

vous

-0.66

osi

-0.65

eph

-0.64

 Meier

-0.64

sm

-0.63

 Amos

-0.62

á

-0.60

ventory

-0.59

fr

-0.58

POSITIVE LOGITS

 training

0.87

 Peaks

0.84

 guiActiveUn

0.82

 sessions

0.80

GGGGGGGG

0.80

 Training

0.79

Training

0.79

 exercises

0.78

 instructors

0.78

 workouts

0.78

Activations Density 0.015%

references to various forms of training

references to training programs or sessions

No Comments

No Known Activations