INDEX

Explanations

occurrences of words representing ordinal positions

oai_token-act-pair · gpt-3.5-turbo

occurrences of the term "second" and related ordinal indicators

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

Tokens

-0.78

rooms

-0.78

maps

-0.71

itu

-0.68

coins

-0.65

tests

-0.65

POST

-0.64

bos

-0.63

auna

-0.62

files

-0.62

POSITIVE LOGITS

 consecutive

0.92

 largest

0.84

 baseman

0.83

 busiest

0.81

 holiest

0.80

 fastest

0.78

 oldest

0.77

 longest

0.77

 highest

0.77

 smartest

0.77

Activations Density 0.059%

occurrences of words representing ordinal positions

occurrences of the term "second" and related ordinal indicators

No Comments

No Known Activations