INDEX

Explanations

negative connotations or implications related to emotional states or situations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res_post_32k-oai

Configuration

jbloom/GPT2-Small-OAI-v5-32k-resid-post-SAEs/v5_32k_layer_6.pt

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

32,768

Data Type

torch.float32

Hook Name

blocks.6.hook_resid_post

Hook Layer

Architecture

standard

Context Size

Dataset

Skylion007/openwebtext

Activation Function

topk

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

eers

-0.80

QC

-0.75

acha

-0.70

=-=-=-=-

-0.70

 DRAG

-0.68

 chants

-0.66

Ń·

-0.65

enegger

-0.64

 nexus

-0.63

ļéĨĴ

-0.63

POSITIVE LOGITS

dozen

1.11

heartedly

1.04

assed

1.00

hearted

0.92

hour

0.89

century

0.89

inning

0.88

assing

0.85

formed

0.85

closed

0.79

Activations Density 0.035%

negative connotations or implications related to emotional states or situations

No Comments

No Known Activations