INDEX

Explanations

terms related to recruitment and professional appointments

oai_token-act-pair · gpt-3.5-turbo

words related to recruitment and recycling processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

inki

-0.79

 ivory

-0.67

drm

-0.62

wa

-0.59

silver

-0.58

 anth

-0.57

oteric

-0.57

 smack

-0.57

 Benz

-0.56

 Tsarnaev

-0.56

POSITIVE LOGITS

rocal

1.14

blance

0.89

issance

0.88

oir

0.85

cled

0.84

facing

0.84

entials

0.84

ction

0.83

lation

0.82

ivism

0.78

Activations Density 0.060%

terms related to recruitment and professional appointments

words related to recruitment and recycling processes

No Comments

No Known Activations