INDEX

Explanations

references to validity or legitimacy 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 @ 8-res_post_32k-oai

Configuration

jbloom/GPT2-Small-OAI-v5-32k-resid-post-SAEs/v5_32k_layer_8.pt

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

32,768

Data Type

torch.float32

Hook Name

blocks.8.hook_resid_post

Hook Layer

Architecture

standard

Context Size

Dataset

Skylion007/openwebtext

Activation Function

topk

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

xual

-0.72

hedon

-0.70

Kut

-0.63

peak

-0.60

 Primordial

-0.60

mania

-0.59

Mania

-0.58

hedral

-0.57

hiba

-0.57

icho

-0.57

POSITIVE LOGITS

ators

1.38

ating

1.38

ator

1.26

ates

1.13

ations

1.10

ated

0.98

ation

0.97

alties

0.94

ifiers

0.92

iation

0.91

Activations Density 0.062%

references to validity or legitimacy in various contexts

No Comments

No Known Activations