INDEX

Explanations

keywords related to programming and coding

oai_token-act-pair · gpt-3.5-turbo

occurrences of instructions or notifications within a technical context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

stre

-0.83

 therap

-0.76

 paternity

-0.73

 Newsp

-0.69

 nurs

-0.67

Cele

-0.67

âĢİ

-0.66

 drowned

-0.65

 Pill

-0.65

 tsun

-0.64

POSITIVE LOGITS

 output

0.97

ython

0.95

 Output

0.95

Output

0.95

 configure

0.88

 python

0.88

 Python

0.84

 default

0.82

output

0.82

 outputs

0.82

Activations Density 0.507%

keywords related to programming and coding

occurrences of instructions or notifications within a technical context

No Comments

No Known Activations