INDEX

Explanations

human mental health-related terms

oai_token-act-pair · gpt-3.5-turbo

word endings that imply a state of being or a condition

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

mic

-0.75

bern

-0.72

mega

-0.71

²¾

-0.68

ãĥ£

-0.67

CHR

-0.66

ilers

-0.65

essen

-0.63

terson

-0.61

 slic

-0.60

POSITIVE LOGITS

naire

1.04

ally

1.03

ment

0.96

ary

0.92

aler

0.90

ateral

0.87

ion

0.87

aires

0.86

alities

0.82

ality

0.80

Activations Density 0.044%

human mental health-related terms

word endings that imply a state of being or a condition

No Comments

No Known Activations