INDEX

Explanations

phrases indicating a point, opinion, or direction

oai_token-act-pair · gpt-3.5-turbo

the preposition "to" 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 @ 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

awei

-0.64

 packets

-0.64

need

-0.62

 snapshots

-0.62

 filters

-0.60

 dips

-0.60

hops

-0.60

 pauses

-0.60

pins

-0.59

 deductions

-0.59

POSITIVE LOGITS

ilet

1.04

asty

1.04

 reiterate

0.97

wered

0.97

ffee

0.95

asted

0.94

othy

0.93

ggles

0.93

pless

0.88

psy

0.87

Activations Density 0.149%

phrases indicating a point, opinion, or direction

the preposition "to" in various contexts

No Comments

No Known Activations