INDEX

Explanations

words related to public health

oai_token-act-pair · gpt-3.5-turbo

mentions of public health organizations and related topics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

*/(

-0.76

coni

-0.74

ptions

-0.71

uble

-0.69

ãĤ§

-0.67

 sadd

-0.67

ãĥĩãĤ£

-0.67

ounter

-0.66

Â¯

-0.65

 slee

-0.64

POSITIVE LOGITS

 Care

0.94

osate

0.94

 Sciences

0.88

care

0.88

hazard

0.86

Care

0.82

 care

0.79

iencies

0.78

aceutical

0.77

iest

0.76

Activations Density 0.033%

words related to public health

mentions of public health organizations and related topics

No Comments

No Known Activations