INDEX

Explanations

phrases related to comparisons or opinions

oai_token-act-pair · gpt-3.5-turbo

phrases that introduce examples or justifications

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

ptives

-0.71

"],

-0.71

ritz

-0.67

taboola

-0.66

aley

-0.66

 Cosponsors

-0.63

ounter

-0.62

 Kirin

-0.61

atform

-0.61

ombat

-0.61

POSITIVE LOGITS

ĪĴ

0.97

 paraph

0.61

acea

0.58

opsis

0.58

 rightly

0.57

mitt

0.57

 parentheses

0.57

pse

0.57

-)

0.56

 bearer

0.55

Activations Density 0.324%

phrases related to comparisons or opinions

phrases that introduce examples or justifications

No Comments

No Known Activations