INDEX

Explanations

programming code snippets with specific keywords and syntax

oai_token-act-pair · gpt-3.5-turbo

occurrences of underscore-prefixed terms related to code or data structures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

quished

-0.85

 Leilan

-0.83

 Manson

-0.77

atform

-0.77

eteria

-0.76

JPM

-0.74

 Pearce

-0.74

 Garry

-0.73

 Doctors

-0.73

 Mercy

-0.73

POSITIVE LOGITS

dict

1.08

chance

1.03

gradient

1.03

blank

1.01

default

0.99

state

0.99

location

0.98

enabled

0.98

token

0.96

base

0.96

Activations Density 0.022%

programming code snippets with specific keywords and syntax

occurrences of underscore-prefixed terms related to code or data structures

No Comments

No Known Activations