INDEX

Explanations

references to doors and their states (open or closed)

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res_post_32k-oai

Configuration

jbloom/GPT2-Small-OAI-v5-32k-resid-post-SAEs/v5_32k_layer_10.pt

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

32,768

Data Type

torch.float32

Hook Name

blocks.10.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

Ranked

-0.91

tein

-0.72

-0.70

abase

-0.69

Own

-0.69

avorite

-0.66

article

-0.65

ry

-0.63

Saharan

-0.62

ombat

-0.61

POSITIVE LOGITS

 wider

1.06

 behind

1.06

 abruptly

1.01

 wide

0.99

 revealing

0.94

 tight

0.93

 tightly

0.90

 softly

0.87

 inward

0.84

 omin

0.83

Activations Density 0.430%

references to doors and their states (open or closed)

No Comments

No Known Activations