INDEX

Explanations

terms related to distribution processes

oai_token-act-pair · gpt-3.5-turbo

terms related to distribution and distributors in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

 Pound

-0.69

laus

-0.67

ever

-0.66

hart

-0.65

icial

-0.62

chan

-0.61

tes

-0.61

Nou

-0.61

pei

-0.61

thur

-0.60

POSITIVE LOGITS

 distributor

1.06

 Distribut

0.97

 distribut

0.96

 distributors

0.93

 receipts

0.85

 warranties

0.80

ãĥ¼ãĥĨãĤ£

0.77

 wholes

0.75

 distribute

0.74

 coupons

0.72

Activations Density 0.009%

terms related to distribution processes

terms related to distribution and distributors in various contexts

No Comments

No Known Activations