INDEX

Explanations

references to failures or shortcomings

oai_token-act-pair · gpt-3.5-turbo

phrases related to failures and responsibilities in various contexts

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

glomer

-0.58

 Sabha

-0.56

 corrid

-0.53

FIR

-0.51

 Kits

-0.50

 wiser

-0.50

 Airl

-0.49

 subdiv

-0.48

IFIED

-0.48

 aspir

-0.48

POSITIVE LOGITS

lessness

0.98

liness

0.85

iveness

0.84

 toward

0.83

iness

0.83

worthiness

0.82

 towards

0.81

fulness

0.79

ahime

0.75

ability

0.74

Activations Density 0.182%

references to failures or shortcomings

phrases related to failures and responsibilities in various contexts

No Comments

No Known Activations