INDEX

Explanations

words related to comparison or contrast

oai_token-act-pair · gpt-3.5-turbo

phrases that introduce a clause or detail, specifically the word "which."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

BLE

-0.62

Behind

-0.61

leton

-0.61

politics

-0.59

ve

-0.59

athi

-0.59

Untitled

-0.58

STE

-0.58

perty

-0.57

selling

-0.57

POSITIVE LOGITS

 translates

1.23

 means

1.20

 begs

1.14

 implies

1.14

 explains

1.13

 brings

1.06

 prompts

1.06

 leads

1.05

 incidentally

1.05

 includes

1.04

Activations Density 0.113%

words related to comparison or contrast

phrases that introduce a clause or detail, specifically the word "which."

No Comments

No Known Activations