INDEX

Explanations

phrases related to trends or changes over time

oai_token-act-pair · gpt-3.5-turbo

conditional and causal relationships in statements

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

ommel

-0.75

bies

-0.69

Fax

-0.69

PASS

-0.67

VICE

-0.67

 chalk

-0.66

 mosqu

-0.66

 impecc

-0.66

anka

-0.66

ocl

-0.66

POSITIVE LOGITS

Yen

0.75

 Slowly

0.69

gradient

0.69

 metast

0.66

 Decay

0.65

luster

0.63

 reorgan

0.63

Reloaded

0.62

 Recession

0.62

 cannibal

0.62

Activations Density 0.801%

phrases related to trends or changes over time

conditional and causal relationships in statements

No Comments

No Known Activations