INDEX

Explanations

instances of the word "at" indicating specific locations or times

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs49152-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

Data Type

torch.float32

Hook Point

blocks.8.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

FTWARE

-1.07

PLA

-0.64

SHIP

-0.61

peria

-0.60

HAM

-0.59

EVA

-0.58

Winged

-0.56

 WRITE

-0.56

Reviewer

-0.55

Ts

-0.55

POSITIVE LOGITS

 least

0.92

roph

0.77

 variance

0.77

 essence

0.75

yp

0.75

opic

0.70

itud

0.69

 times

0.68

eatures

0.67

vet

0.67

Activations Density 0.059%

instances of the word "at" indicating specific locations or times

No Comments

No Known Activations