INDEX

Explanations

code snippets involving loops and flow control statements

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

programming concepts related to iteration and control flow

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 12-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.hook_resid_post

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.11.hook_resid_post

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

 UNESCO

-0.68

 Official

-0.67

CTV

-0.66

 Darkness

-0.64

 Shame

-0.63

CLASSIFIED

-0.62

 Darling

-0.61

apolog

-0.61

 Sturgeon

-0.60

 Underground

-0.60

POSITIVE LOGITS

len

1.00

len

1.00

index

0.98

 indices

0.97

0.94

length

0.91

xs

0.91

seq

0.91

Index

0.90

 iter

0.90

Activations Density 0.078%

code snippets involving loops and flow control statements

programming concepts related to iteration and control flow

No Comments

No Known Activations