INDEX

Explanations

phrases related to specific technical terms or jargon

oai_token-act-pair · gpt-3.5-turbo

words related to specific categories or classifications

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

Rap

-0.72

!".

-0.67

Cash

-0.67

]."

-0.62

inki

-0.60

Charge

-0.58

.ï¿½

-0.58

abwe

-0.57

çĶ°

-0.57

APD

-0.57

POSITIVE LOGITS

endum

0.61

emn

0.58

ritical

0.57

reth

0.54

gins

0.53

pedia

0.53

uous

0.53

 pane

0.53

enary

0.51

ship

0.51

Activations Density 1.153%

phrases related to specific technical terms or jargon

words related to specific categories or classifications

No Comments

No Known Activations