INDEX

Explanations

commands or instructions within a text

oai_token-act-pair · gpt-3.5-turbo

references to instructions or actions that involve a sequence of steps or commands

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

±

-0.82

¶æ

-0.79

aukee

-0.77

DERR

-0.73

ibles

-0.69

)</

-0.68

emy

-0.67

gran

-0.64

gger

-0.64

uers

-0.64

POSITIVE LOGITS

:-

0.87

:(

0.87

 assumes

0.76

0.75

>:

0.72

 illustrates

0.71

 configure

0.70

 summarizes

0.70

 demonstrates

0.69

redits

0.68

Activations Density 0.081%

commands or instructions within a text

references to instructions or actions that involve a sequence of steps or commands

No Comments

No Known Activations