INDEX

Explanations

lines of code written in a specific programming language

oai_token-act-pair · gpt-3.5-turbo

programming-related symbols and assignment statements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

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

itant

-0.87

keeper

-0.85

grounds

-0.84

soDeliveryDate

-0.80

arians

-0.79

forth

-0.79

itage

-0.74

ogyn

-0.74

ened

-0.74

prising

-0.73

POSITIVE LOGITS

[];

1.23

{}

1.17

========

1.12

1.09

============

1.05

{"

1.05

[|

1.03

 lambda

1.02

[]

0.96

"{

0.94

Activations Density 0.028%

lines of code written in a specific programming language

programming-related symbols and assignment statements

No Comments

No Known Activations