INDEX

Explanations

mentions of mental health conditions

oai_token-act-pair · gpt-3.5-turbo

variations of the word "ment," indicating a focus on concepts or processes related to development and mental states

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

mic

-0.74

éĹĺ

-0.68

soDeliveryDate

-0.67

CHR

-0.67

 uncontroll

-0.62

 mosqu

-0.61

ECD

-0.61

ilers

-0.61

pins

-0.61

terson

-0.60

POSITIVE LOGITS

ally

1.09

ment

1.05

aler

0.97

ateral

0.96

ion

0.92

ary

0.89

naire

0.89

ioned

0.82

ists

0.82

ments

0.81

Activations Density 0.031%

mentions of mental health conditions

variations of the word "ment," indicating a focus on concepts or processes related to development and mental states

No Comments

No Known Activations