INDEX

Explanations

adjectives indicating uncertainty or perception

oai_token-act-pair · gpt-3.5-turbo

instances of the word "seemingly" to highlight uncertain or ambiguous situations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 Regions

-0.78

ioch

-0.76

ription

-0.75

ests

-0.72

ogie

-0.68

awks

-0.68

ogi

-0.67

ablishment

-0.65

alez

-0.64

 Drill

-0.64

POSITIVE LOGITS

 innocuous

1.10

 unstoppable

1.04

 oblivious

0.99

 unrelated

0.97

 contradictory

0.94

 endless

0.92

 limitless

0.90

 unaware

0.90

 insur

0.90

 unbeat

0.81

Activations Density 0.050%

adjectives indicating uncertainty or perception

instances of the word "seemingly" to highlight uncertain or ambiguous situations

No Comments

No Known Activations