INDEX

Explanations

references to legal cases or incidents reported in news articles

oai_token-act-pair · gpt-3.5-turbo

references to legal or criminal cases

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

asus

-0.87

 livest

-0.79

 embr

-0.71

oir

-0.70

owered

-0.70

itage

-0.70

atomic

-0.70

 mathemat

-0.69

anian

-0.69

jri

-0.68

POSITIVE LOGITS

 involving

1.06

cases

0.89

 cases

0.89

0.74

esan

0.73

cale

0.73

 diagnosed

0.69

 warrants

0.67

 cited

0.66

cmp

0.65

Activations Density 0.030%

references to legal cases or incidents reported in news articles

references to legal or criminal cases

No Comments

No Known Activations