INDEX

Explanations

numbers that are part of news headlines

oai_token-act-pair · gpt-3.5-turbo

numerical values and references to events or measurements

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

imov

-0.69

hip

-0.68

ument

-0.67

pen

-0.67

esse

-0.66

ANE

-0.66

cot

-0.65

ropri

-0.65

PI

-0.63

enne

-0.63

POSITIVE LOGITS

 darts

0.67

terms

0.65

 rounds

0.63

 wors

0.63

 stairs

0.62

prints

0.62

 branches

0.61

 Languages

0.61

 Tests

0.60

è¦ļéĨĴ

0.60

Activations Density 0.036%

numbers that are part of news headlines

numerical values and references to events or measurements

No Comments

No Known Activations