INDEX

Explanations

specific terms related to bras

oai_token-act-pair · gpt-3.5-turbo

references to bras

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

 Procedure

-0.80

 Trafford

-0.67

 Chronicle

-0.67

administ

-0.65

eers

-0.63

 Milton

-0.63

 Memor

-0.62

 PowerPoint

-0.62

irrel

-0.61

 Fiction

-0.61

POSITIVE LOGITS

ided

1.07

 straps

0.92

 bras

0.91

 strap

0.90

ille

0.90

loads

0.87

ignt

0.87

Ã§

0.84

zzle

0.84

cies

0.84

Activations Density 0.013%

specific terms related to bras

references to bras

No Comments

No Known Activations