INDEX

Explanations

instances of the word "at" used in various contexts

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

SELECT

-0.76

soDeliveryDate

-0.75

margin

-0.72

Label

-0.72

ahime

-0.71

positive

-0.69

Thumbnail

-0.69

UP

-0.68

GREEN

-0.67

PER

-0.67

POSITIVE LOGITS

 least

1.18

 passers

0.86

him

0.82

 unsuspecting

0.82

roph

0.79

 fleeing

0.77

 them

0.76

 random

0.76

onement

0.74

 unarmed

0.74

Activations Density 0.044%

instances of the word "at" used in various contexts

No Comments

No Known Activations