INDEX

Explanations

information related to specific pieces of writing, such as articles, blog posts, letters, and reports

oai_token-act-pair · gpt-3.5-turbo

clauses that introduce significant details about events or statements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 accommodation

-0.70

pants

-0.67

Course

-0.62

tsy

-0.61

 wear

-0.61

unts

-0.61

animate

-0.60

 seas

-0.60

mith

-0.60

hem

-0.59

POSITIVE LOGITS

itled

1.16

vertisement

1.07

 titled

0.98

 headlined

0.95

 excerpts

0.94

uggest

0.89

 outlining

0.86

lished

0.85

 detailing

0.85

 entitled

0.84

Activations Density 0.484%

information related to specific pieces of writing, such as articles, blog posts, letters, and reports

clauses that introduce significant details about events or statements

No Comments

No Known Activations