INDEX

Explanations

words related to adding or including something new

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "added" in various contexts

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

bin

-0.74

falls

-0.72

wh

-0.67

Ago

-0.66

view

-0.64

ARE

-0.64

ograms

-0.64

Bos

-0.64

bia

-0.64

 Â¯

-0.63

POSITIVE LOGITS

endum

1.01

itionally

0.98

added

0.91

ictions

0.88

 insult

0.86

itional

0.82

 thereto

0.82

ition

0.82

itions

0.79

itivity

0.78

Activations Density 0.038%

words related to adding or including something new

occurrences of the word "added" in various contexts

No Comments

No Known Activations