INDEX

Explanations

words related to programming and coding concepts mentioned in instructional text

oai_token-act-pair · gpt-3.5-turbo

the presence of the word "is."

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

apixel

-0.63

 lapt

-0.63

 Spani

-0.62

 toile

-0.61

 sacrific

-0.59

åº

-0.59

 precinct

-0.58

 Prior

-0.58

 slab

-0.57

 Brune

-0.57

POSITIVE LOGITS

pecially

0.82

etsk

0.78

outhern

0.76

ELF

0.75

 ironic

0.75

ashtra

0.72

forth

0.72

 unclear

0.72

fits

0.72

 conceivable

0.70

Activations Density 0.123%

words related to programming and coding concepts mentioned in instructional text

the presence of the word "is."

No Comments

No Known Activations