INDEX

Explanations

the phrase "in other words" indicating further explanation

oai_token-act-pair · gpt-3.5-turbo

phrases that introduce or summarize information

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

Ã±o

-0.72

rolog

-0.68

ramid

-0.66

©¶æ¥µ

-0.65

cients

-0.64

onite

-0.62

roid

-0.62

arp

-0.62

rob

-0.61

rift

-0.61

POSITIVE LOGITS

mith

1.35

terday

1.10

 words

0.93

sworth

0.85

words

0.82

 aloud

0.82

paces

0.76

 guiActiveUn

0.72

word

0.71

poons

0.71

Activations Density 0.012%

the phrase "in other words" indicating further explanation

phrases that introduce or summarize information

No Comments

No Known Activations