INDEX

Explanations

words related to specific locations or establishments, particularly those starting with 'St'

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs1536-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

1,536

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ļéĨĴ

-0.75

hower

-0.73

 "$:/

-0.69

hound

-0.66

EStream

-0.61

 limited

-0.61

 thumbs

-0.59

 prompt

-0.59

 friendly

-0.58

 deaf

-0.58

POSITIVE LOGITS

rict

1.20

alker

1.14

onew

1.10

amped

1.09

uffed

1.08

oppable

1.07

itched

1.06

roller

1.06

okes

1.06

omach

1.04

Activations Density 1.104%

words related to specific locations or establishments, particularly those starting with 'St'

No Comments

No Known Activations