INDEX

Explanations

programming-related syntax and keywords

oai_token-act-pair · gpt-3.5-turbo

technical terms related to coding errors and compilation processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

uran

-0.74

ayne

-0.68

rotein

-0.63

arian

-0.62

oths

-0.61

TBA

-0.60

aways

-0.60

 Watts

-0.59

 Cheong

-0.59

attled

-0.59

POSITIVE LOGITS

XXX

0.95

xxx

0.88

foo

0.86

[]

0.84

example

0.83

Debug

0.83

XXXX

0.82

xxxx

0.81

foo

0.81

default

0.79

Activations Density 0.317%

programming-related syntax and keywords

technical terms related to coding errors and compilation processes

No Comments

No Known Activations